Dilemmas in the treatment of early-onset first-episode psychosis

Abstract

Early-onset first-episode psychosis (EOP) is a severe mental disorder that can pose a number of challenges to clinicians, young people and their families. Its assessment and differentiation from other neurodevelopmental and mental health conditions may at times be difficult, its treatment may not always lead to optimal outcomes and can be associated with significant side effects, and its long-term course and prognosis seem to be less favourable compared with the adult-onset disorder. In this paper, we discuss some dilemmas associated with the evaluation and management of EOP and propose approaches that can be used in the clinical decision-making process. A detailed and well-informed assessment of psychotic symptoms and comorbidities, a systematic approach to treatment with minimum possible medication doses and close monitoring of its effectiveness and adverse effects, and multidimensional interventions taking into consideration risks and expectations associated with EOP, are paramount in the achievement of the most favourable outcomes for affected children and young people.

Introduction

Psychosis is an umbrella term for a clinical presentation conceptualized, both by the American Psychiatric Association¹ and the World Health Organization,² as a combination of hallucinations, delusions, disorganized thinking or behaviour, negative or catatonic symptoms, and functional impairment.³ In the current diagnostic classification systems, impaired reality testing remains its hallmark. Psychotic symptoms are the essential feature of schizophrenia spectrum disorders, but may also be present in mood disorders,^4–6 autism spectrum disorders,⁷ personality disorders,^8,9 substance misuse^6,10 and in the normal population.^11,12 Psychosis may add to impairment and disability,¹³ affect cooperation with treatment¹⁴ and increase stigma associated with mental health difficulties.¹⁵ The evaluation of psychotic symptoms and identification and treatment of psychosis are, therefore, important targets of the assessment and intervention of patients seen by the mental health services.

The current prevailing explanatory model for psychosis, mainly developed through studies of schizophrenia, points towards the condition being neurodevelopmental in origin, with the interaction between genetic and environmental factors marking the beginning of brain pathophysiological processes long before the overt manifestation of clinical symptoms. Several lines of evidence suggest that psychosis is more severe, is associated with more neurodevelopmental deviance and possibly has poorer prognosis, the earlier it presents.^16–19 Within this framework, early-onset psychosis (EOP) has received particular attention as it provides on the one hand a unique opportunity to explore the aberrant neurodevelopmental origins of the condition and on the other hand, potential opportunities to affect its long-term course. Despite research into the phenomenology, epidemiology, pathophysiology and treatment options in EOP, there are several challenges that clinicians face in the assessment and management of children and young people presenting with psychotic symptoms for the first time. In this paper, we will discuss some of the dilemmas related to the treatment of psychosis presenting for the first time in children and adolescents and ways to navigate through these (Table 1).

Table 1.

Dilemmas in the treatment of early-onset first episode psychosis.

(1) What are we treating?

(a) Early-onset psychotic disorder is rare but psychotic symptoms are frequent.

(b) Frequent comorbidity and atypical presentation can complicate diagnosis.

(c) Developmental and environmental factors are very relevant.

(d) ‘Narrower’ approach to psychosis as a disorder is recommended.

(2) How do we treat?

(a) Second-generation antipsychotics are the treatment of choice.

(b) Side-effect profile guides treatment choice, given the similar effectiveness of different SGAs.

(c) Cautious initiation and gradual increase of the dose of antipsychotic should be applied.

(d) Clozapine should be considered in treatment-resistant cases.

(e) Long-term treatment is likely to be needed.

(f) Psychological interventions should be offered.

(3) What should we expect from treatment?

(a) Resolution of symptoms should be the aim of treatment but prognosis is poorer compared with adult-onset cases.

(b) Treatment of comorbidities is an important aspect of long-term management.

(c) Suicide prevention is paramount.

(d) Adverse effects of medication should be minimized.

SGAs, second-generation antipsychotics.

Dilemma number 1: what are we treating?

The assessment of psychotic disorders can be complicated, and its challenges can be exacerbated by the precocious onset of these conditions. Psychosis presenting in childhood and adolescence, whilst to a large extent characterized by the same combination of symptoms as with the adult-onset disorder, can manifest in ways that are more easily subject to alternative diagnostic formulations. Only a small percentage of psychosis cases are manifest before adulthood,^20,21 and presentations can be complicated by comorbidities²² which may both serve as vulnerability for the evolution of the disorder, but also alter its clinical picture. Furthermore, EOP is often characterized by a more protean presentation, which can be harder to differentiate from combinations of neurodevelopmental (e.g. autism spectrum disorders) with mental health conditions (e.g. anxiety or mood disorders).^7,23 In addition, given the relative infrequency of psychosis in adolescents and the extreme rarity in children, more comprehensive physical investigations need to be undertaken, depending on the age at onset and clinical presentation, in order to exclude neuropsychiatric complications of physical conditions.⁷ The nature of psychotic experiences in childhood and adolescence is also subject to differences of interpretation, with evidence of childhood hallucinatory experiences frequently indexing emotional and conduct problems rather than a psychotic disorder.²⁴ The prevalence of unusual experiences in clinical populations of this age group has been identified as very high. In a recent study, direct evaluation through self report of psychotic-like symptoms experienced by children and adolescents attending community mental health clinics was found to reach 68%, with 60% of these being associated with distress or impairment.²⁵ Finally, the multiple routes into psychosis, affected both by genetic liability²⁶ and psychosocial adversity and abuse²⁷ may not always provide a good pointer towards EOP as they are not unique to psychosis, also being associated with a wide range of brain changes and mental and behavioural disorders with commonly adolescent onset.^28–30 The interaction between brain maturational processes, environmental insults and proneness to mental illness are more likely to give mixed or hard-to-deconstruct clinical presentations in this age when EOP incidence gradually reaches its peak.

Unravelling this diagnostic dilemma is crucial because the provision of early, robust and effective treatment relies on accurate diagnosis. When diagnostic clarity does occur, there is evidence of good diagnostic stability over time for early-onset schizophrenia,^31,32 and bipolar disorder³² but not as good for psychosis associated with other conditions, for example, depressive disorder.³² It is therefore very important to avoid premature diagnostic foreclosure in the face of uncertainty. The evolution of psychosis does go through stages; functional impairment in the prodrome can be interpreted as affective disorder, or even a manifestation of normal adolescence, and a detailed developing assessment and staged approach to treatment interventions are warranted. Careful consideration of developmental history and environmental circumstances, characterization of psychotic symptoms, comprehensive evaluation for mood and anxiety disorders, exclusion of organic conditions and substance misuse, and conclusion about the degree of impairment associated with psychotic symptoms are likely to facilitate the diagnostic formulation. A balance between indiscriminate classification of psychotic symptoms as a psychotic disorder and reluctance to give the diagnosis needs to be maintained. In early-onset cases, the accurate characterization of the young person’s difficulties, and a narrower approach to psychosis as a disorder to include a more severe and impairing combination of psychotic symptoms is likely to be the most appropriate, clinically, to identify evidence-based interventions.

Dilemma number 2: how do we treat?

In children and young people, once the diagnosis of psychosis is made, second-generation antipsychotics (SGAs) are generally recommended.³³ Although the choice of treatment is hampered by some limitations in evidence base, both first- and second-generation antipsychotics are of established and comparable efficacy in psychosis but differ in their side-effect profile.^34,35 Use of SGAs is almost ubiquitous, but perhaps needs to be rebalanced in context of patient discussion about acceptability of side effects, both in the short and longer term, given the likelihood of necessity for continuing treatment. First-generation antipsychotics are generally to be avoided due to their side-effect profile, in particular, high incidence of movement disorders, including tardive dyskinesia.³⁶ The safety of antipsychotics in childhood and adolescence remains a key area of clinical uncertainty, and risk/benefit analysis, mostly favouring their use in severe mental illness, needs to be undertaken in all cases.³⁵ The choice of antipsychotic is to a large extent determined by its side-effect profile. Of all the SGAs, olanzapine is associated with more weight gain and metabolic side effects, so when this choice is made, it needs to be discussed with young people and their families.³³ Most clinicians would not prescribe olanzapine routinely as first line in EOP. Additional treatments may also be prescribed to target comorbid conditions, for example, mood stabilisers in presentations with a bipolar affective component or selective serotonin reuptake inhibitors for comorbid anxiety and depression. There are a growing number of trials in this population reflecting the importance of generating a child-specific evidence base for prescribing antipsychotics. To that end, international initiatives to set standards and optimise the methodological and practical aspects of research in children, such as the Standards for Research in Children (STaR Child Health) and the Child and Adolescent Psychiatry Trials Network (CAPTN) group have been developed.^37,38

The only antipsychotic which seems to be more efficacious compared with the rest of the SGAs is clozapine.^39–41 Although its superiority in meta-analyses of adult samples has recently been a subject of debate,^42,43 it was also argued that trials including children and young people need to be considered separately.⁴⁴ The evidence base on clozapine in this age group is very limited and clozapine trials are generally difficult to conduct due to the small number of cases that can potentially be included in them. Clozapine remains an underused drug across the lifespan^40,45 that results in treatment-resistant cases going through repeated cycling through pharmacopoeia rather than having clozapine tried. This is likely due to barriers to acceptability, and whilst risk issues can be a concern for patients and families, professional difficulties also compromise its use with many child and adolescent psychiatrists being wary of the drug due to unfamiliarity outside of inpatient units. This situation may also be perpetuated by lack of resources to initiate clozapine in the community in this population and services not being set up to manage the monitoring processes required. However, in cases not responding to alternative SGAs, clinicians should become more comfortable considering this option as the majority of EOP patients who are eventually prescribed clozapine appear to have a favourable clinical outcome.⁴⁰

In this younger age group, the principle of cautious initiation and gradual increase of the dose of antipsychotic is clinically applied, but needs to be balanced against the risk of delayed or inadequate treatment.⁴⁶ Whilst the principle of ‘start low, go slow’ is widely employed, young people are not uniquely responsive to the beneficial effects of medication, but rather sensitive to side effects that can impair tolerability and compliance.⁴⁷ The decision about when an antipsychotic is considered to not be effective has also been subject to investigation, with some evidence suggesting that this may be concluded as early as 2–3 weeks into treatment with an adequate dose of the drug.^48,49 However, taking the current guidelines into account, it is important to consider that early discontinuation may affect eventual classification of cases as treatment resistant,⁵⁰ which is likely to have implications for clozapine use. In addition to the absence of clear treatment response, there can also be uncertainty in relation to suboptimal response, problems with tolerability or a combination of the two. Although these challenges apply similarly to some extent to adults, ongoing lack of capacity or competence in children and young people make such discussions with the wider family more complicated. Alongside clinical judgement and close physical health monitoring, the use of rating scales to quantify the young person’s improvement and side effects are likely to assist with these decisions.^7,51

Alternative forms of administration, for example, long-acting injectable (LAI) medication, have a role in the management of young people when there are concerns about nonadherence, but sometimes clarity about the necessity for using such interventions only becomes apparent in retrospect. Nonadherence certainly is an issue with adolescents, not just with regard to psychotropic medication,⁵² but also for other diseases requiring long-term pharmacological intervention e.g. diabetes⁵³ or epilepsy.⁵⁴ This may be embedded in developmental processes around autonomy and control,⁵⁵ and difficulties accepting long-term consequences of current difficulties.⁵⁶ Nevertheless, when LAIs are considered for EOP, due to serious adherence issues and significant risks associated with the young people’s clinical presentation, careful monitoring is warranted due to almost complete lack of trial evidence in this age group. It is encouraging that, in adult studies, LAIs were similar to oral antipsychotics regarding the frequency of treatment discontinuation due to adverse events and serious adverse events. However, their use was associated with significantly more akinesia, low-density lipoprotein cholesterol changes and anxiety but also, interestingly, with significantly lower prolactin change.⁵⁷

Treatment with antipsychotics in schizophrenia spectrum disorders is usually long term. No clear evidence exists about when it may be safe to reduce the dose or discontinue medication following the first episode of EOP and, from a practical perspective, such approaches may only be discussed in the case of full remission for a very considerable amount of time. Under these circumstances, potential reduction of medication needs to be very gradual and be accompanied with very close monitoring of its outcome. With the first signs of clinical deterioration, medication needs to be reinstated as previously, to avoid another episode of psychosis. The development of side effects may also trigger discussions about medication discontinuation or switching to an alternative antipsychotic. Only one study to date⁵⁸ has investigated antipsychotic withdrawal in adolescents with schizophrenia using a double-blind placebo-control randomized design. In this study, patients were cross titrated and stabilized from their antipsychotic treatment to aripiprazole, and then randomized to aripiprazole or placebo in a 52-week double-blind maintenance phase. Compared with placebo, active treatment was associated with a significantly longer time to exacerbation of psychotic symptoms or impending relapse, lower rates of serious or severe adverse events, lower rate of discontinuation due to serious or severe adverse events, and similar or lower incidences of extrapyramidal symptoms, weight gain, and somnolence.

Although antipsychotic medication is the first-line treatment for EOP, psychological interventions also have a role to play in its management. Psychological therapies need to be adapted for children and adolescents with psychosis.³³ These adaptations take into consideration cognitive, social, psychological and practical aspects specific to their age. From a practical perspective, compared with mood and anxiety disorders, young people with EOP seek psychological support less often and their engagement may not be as good. They need a flexible and creative approach; one that adapts to their understanding and the language they use about their experiences. The evidence base for psychological therapies specific to adolescent psychosis is very limited and there is practically no evidence for their use in children. Cognitive remediation therapy (CRT) seems to be beneficial for improving cognitive ability in EOP^59–61 and also psychosocial functioning⁵⁹ with the latter also being shown to improve by a combination of cognitive behavioural therapy (CBT) and family intervention in a small pilot study.⁶² Studies focusing on EOP that examined therapy effects on psychotic symptoms have reported negative results.^59,61,62 More research on psychological interventions in this age group is needed.

Dilemma number 3: what should we expect from treatment?

The majority of EOP patients will have further episodes of illness and long-term difficulties. Having at least one further episode requiring hospitalization was reported to reach 83% of adolescents with schizophrenia over a period of more than 10 years⁶³ with the majority of the sample also having at least moderate educational and occupational impairment and serious social disability. Two studies that followed up childhood-onset patients over a period of 42 years^18,64 also reported poor outcomes, with half of them being found to have continuous symptoms, 25% to be in partial remission and less than a sixth showing a favourable outcome. More than 70% did not graduate from school and were unemployed at the time of follow up.¹⁸ Two age cut-off points delineating worse prognosis, as assessed by number of days in first admission, percentage of patients with more than one admission, average number of days in hospital and number of admissions per illness year, seem to be around 17 and 12 years, with patients having earlier onset being more affected.¹⁷

As in all other clinical conditions, the aim of treatment in EOP should always be for symptom resolution, with a return to premorbid levels of functioning. However, this can be difficult to achieve in the context of increased severity and poorer prognosis associated with earlier onset,⁶⁵ so there is an imperative to ensure all remediable issues are dealt with, in terms of prompt effective treatment in multiple modalities.³³ Within EOP, poorer premorbid functioning in childhood, the severity of negative symptoms at baseline and the duration of untreated psychosis are all associated with a worse outcome.⁶⁶ Inferior treatment response to antipsychotic medication has also been associated with comorbid autism spectrum disorders.⁶⁷ Even in the event of a good response to treatment, EOP has a significant impact on young people. The recovery model⁶⁸ has had a significant influence on mental health policy and provision, and whilst its tenets are applicable to young people with EOP, it is important to acknowledge that despite symptomatic recovery, young people are on a chronologically determined trajectory through educational attainments and social transitions that they cannot step out of, in the context of an episode of illness, and return to at the same point, rather than behind similarly aged peers.

A particular concern in relation to long-term management in EOP is suicide. Suicide rates for people with schizophrenia are elevated compared with the general population, with an estimated lifetime risk of approximately 5%.⁶⁹ The strongest associations with later suicide include being young and male, and experiencing a greater burden of illness. In early-onset cases, suicide rates may be much higher; a study following up 61 patients over a period of at least 11 years⁷⁰ identified 9 patients dying of suicide (14.7%) with this being particularly elevated in males (6/28; 21.4%). Considering the possibility of this devastating outcome is directly relevant to risk management in EOP, further emphasizing the necessity for effective treatment for psychosis, including the use of clozapine⁷¹ as a specific suicide-reduction strategy.

Taking into account the increased susceptibility of young people to adverse side effects of medication, a not uncommon clinical dilemma emerges when a good clinical response is associated with a side-effect profile that has significant biological or psychosocial implications, such as excessive weight gain, sexual dysfunction or sedation. From the clinician’s point of view, altering established and effective medication regimes can generate understandable anxiety with regard to risk of relapse. However, given the likelihood for the necessity for long-term treatment, efforts should be made to establish the minimum effective dose, or to switch young people to medication with less propensity to cause side effects of particular salience to them. This not only affects the young person’s quality of life, physical health and long-term morbidity and mortality, but increases the chances of treatment compliance, which is a significant concern in adolescent-onset psychotic disorders, as mentioned above.

Significant dilemmas associated with long-term expectations from treatment also emerge in relation to the management of comorbid conditions. Comorbidities may include neurodevelopmental disorders, like autism spectrum disorders, which affect the clinical presentation of young people with psychosis in remission and will require active management of associated impairment on its own right.⁷ Psychosis in young people is also highly comorbid with depressive and anxiety disorders⁶⁹ which can be harder to identify as they may be considered to be part of the psychosis phenomenology or represent medication side effects and are therefore left untreated. Low expectations of full recovery may affect clinicians’ motivation to diagnose additional conditions and therefore miss the opportunity to implement evidence-based interventions which are likely to significantly improve the clinical progress of the young person. Finally, substance misuse, frequently preceding or coexisting with psychosis,^72,73 can influence not only the aetiological conceptualization of young people’s symptomatology, but also lower the expectations of recovery and lead to suboptimal treatments of both conditions.

Conclusion

First-episode EOP is a severe mental health condition which requires timely specialist intervention. Although our understanding of the significance of psychotic symptoms in clinical and nonclinical populations and our conceptualization of EOP and its overlap with other conditions have significantly evolved in the last few decades, its treatment remains to a large extent suboptimal. Its clinical presentation and management can be associated with significant difficulties, giving rise to dilemmas both for the treating clinician and the young person and their family. Better characterization of its phenomenology, identification and application of the most favourable treatment strategies, and realistic but also well informed expectations are likely to result in the achievement of the best possible clinical outcomes, and assist in the young person reaching their full potential. Further research targeting improved medication effectiveness, optimal use of current treatment options, evaluation of treatment failure, side-effect reduction and the development of psychosocial interventions in children and young people with EOP is urgently needed.