Frequency and Correlates of Acute Dystonic Reactions After Antipsychotic Initiation in 441 Children and Adolescents

Abstract

Objective:

To determine the incidence of acute dystonic reactions (ADRs) and risk factors for ADRs in children and adolescents treated with antipsychotics.

Methods:

This was a retrospective chart review-based cohort study of consecutive patients who attended a university hospital's child and adolescent psychiatry department between 2015 and 2017 and who were treated with antipsychotics and had at least two follow-up visits.

Results:

Thirty of 441 patients (6.8%) 4–19 years of age who were treated with antipsychotics for conduct disorders (21.5%), attention-deficit/hyperactivity disorder (13.2%) and, irritability and aggression that accompanied intellectual disability (12.9%) and followed for 99.5 ± 223.3 (median: 34) days developed ADRs. ADRs developed in 11/391 patients (2.8%) treated with one antipsychotic and 19/50 patients (38.0%) treated with two antipsychotics (p < 0.001). In patients treated with one antipsychotic that developed ADRs, the time to ADRs was 4.0 ± 4.0 days after antipsychotic initiation and 2.7 ± 2.4 days after an increase in the antipsychotic dose. The time to ADRs in those treated with two antipsychotics was 3.0 ± 2.3 days after the addition of the second antipsychotic and 1.6 ± 0.8 days after a dose increase in the second antipsychotic. The incidence of ADRs during antipsychotic monotherapy was 10.5% with first-generation antipsychotics (FGAs) and 2.2% with second-generation antipsychotics (SGAs; p = 0.037). The antipsychotic was changed due to ADRs in 12/30 (40.0%) of ADR cases. Independent factors associated with ADRs were antipsychotic polypharmacy (p < 0.0001), inpatient treatment (p = 0.013), FGA use (p = 0.015), and diagnoses of schizophrenia (p = 0.039) or bipolar disorder (p < 0.0001).

Conclusion:

SGAs and low-potency FGA monotherapy in children and adolescents were associated with a relatively low ADR risk, whereas high- and mid-potency FGAs were associated with a high risk. Independent predictors of ADRs were antipsychotic polypharmacy, inpatient treatment, FGAs, and schizophrenia or bipolar disorder diagnoses, which may be related to more aggressive antipsychotic dosing.

Introduction

Dystonia is an involuntary motor syndrome that is usually spasmodic but may be persistent and involve concomitant contractions of agonist and antagonist muscles (Abdo et al. 2010). Drug-induced dystonic reactions are the most common cause of acute secondary dystonia (Cossu and Colosimo 2017). Antipsychotics are among the most commonly reported causes of drug-related dystonic reactions in children and adolescents (Derinoz and Caglar 2013). The use of antipsychotics, especially second-generation antipsychotics (SGAs), has increased in children and adolescents in recent years (Olfson et al. 2012, 2014; Penfold et al. 2013). The increase in usage may be due to FDA approval of these drugs for children and adolescents with schizophrenia, bipolar disorder, autism-related irritability, Tourette syndrome, and bipolar depression (Christian et al. 2012); as well as their off-label use for other reasons, such as sleep disorders, attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder, major depressive disorder, and posttraumatic stress disorder (Penfold et al. 2013).

Previous studies on extrapyramidal system (EPS) side effects of antipsychotics in children and adolescents reported that neuromuscular problems were less common with SGAs than with first-generation antipsychotics (FGAs) (Satterwaite et al. 2008; Sarkar and Grover 2013). In a randomized study on children and adolescents 8–19 years of age with psychotic symptoms, EPS side effects that required treatment with anticholinergics were reported in 67% of haloperidol users (mean dose: 5.0 ± 2.0 mg/d), 56% of olanzapine users (mean dose: 12.3 ± 3.5 mg/d), and 53% of risperidone (4.0 ± 1.2 mg/d) users (Sikich et al. 2004). Pharmacodynamically, the risk of acute dystonic reactions (ADRs) with SGAs is lower than with FGAs, as SGAs work by either partial D2 agonism or blocking serotonin 5HT2A receptors and dopamine D2 receptors (Correll 2010). However, when the dose of SGAs is sufficiently increased, D2 blockade increases, and EPS side effects can emerge. ADRs can therefore also be observed both in adults (Leucht et al. 2013) and youth (Correll 2008) using SGAs.

In randomized controlled trials, subject/investigator-reported EPS side effect frequencies varied by methodology and population, ranging from 11% to 18% for ziprasidone, 15% to 39% for aripiprazole, 0% to 46% for quetiapine, and 15% to 25% or 0% to 51% for risperidone, depending on the assessment method and dose, and 0% to 58% for olanzapine (Carbon et al. 2015). In a randomized controlled trial, anticholinergic prescriptions, indicating clinically significant Parkinsonism, varied between 12% for olanzapine and 32% for risperidone (Sikich et al. 2008). On the other hand, in a 12-week naturalistic study, anticholinergics were most frequently coprescribed with risperidone (10.2%), followed by aripiprazole (4.8%), and olanzapine, quetiapine, and ziprasidone did not lead to anticholinergic use (Carbon et al. 2015).

ADRs are among the most severe acute EPS side effects and can lead to discomfort and anxiety in terms of antipsychotic use, as well as treatment discontinuation (Pringsheim et al. 2011). For example, in a naturalistic, 12-week study, oropharyngeal acute dystonia developed in 3 of 15 patients using risperidone or aripiprazole, with one of these patients discontinuing the treatment (Menard et al. 2014). In a retrospective chart review study on patients 4–18 years of age, dystonia developed in 1 of 45 patients (2.2%) using risperidone and 1 of 24 (4.1%) patients using olanzapine in the first 3 months of treatment (Alacqua et al. 2008). A review of studies on antipsychotic use in children and adolescents found fewer EPS side effects with SGAs than FGAs (Seida et al. 2012).

Previous data on FGA use indicated that EPS side effects appeared to be more frequent in children and adolescents than in adults (Correll et al. 2006). However, a recent naturalistic study on children and adolescents treated with moderate doses of SGAs, with slow drug titration concluded that EPS side effects did not seem to be more common in this population than in adults (Carbon et al. 2015). The study did not report the incidence of ADRs. To the best of our knowledge, no studies have specifically focused on ADRs in children and adolescents treated with antipsychotics and ADR-related risk factors in sufficiently large samples.

The aim of the present study was to assess the frequency of ADRs and related risk factors in children and adolescents receiving antipsychotic treatment. With this aim, we conducted a retrospective chart review of ADRs in a consecutive cohort of children and adolescents treated with antipsychotics, focusing also on the incidence of ADR development associated with specific antipsychotic agents in this age group.

Methods

Sample

We obtained approval from the Ethics Committee of the Ankara Yildirim Beyazit University Faculty of Medicine, Yenimahalle Training and Research Hospital for this retrospective cohort study. The charts of all consecutive patients who were referred to the Ankara Yildirim Beyazit University's Department of Child and Adolescent Psychiatry between July 1, 2015 and January 1, 2017 and prescribed antipsychotics were examined. The study inclusion criteria were as follows: (1) being between 4 and 19 years of age, (2) being prescribed an antipsychotic, and (3) having minimum two follow-up visits at our hospital. All psychiatric disorders were diagnosed according to Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) criteria (American Psychiatric Association 2013). The treatment characteristics of the cases are presented in Figure 1.

FIG. 1.

The duration of treatment/observation and the timing of the acute dystonic reaction.

S.T.H. and M.F.C. abstracted the data, and discrepancies were resolved by assessing the Udvalg für Kliniske Undersogelser Side Effect Rating Scale (UKU-SERS) of each patient (Lingjaerde et al. 1987).

Data collection tools

Age, sex, clinically obtained DSM-5 diagnoses, antipsychotic name and dose, and side effects reported by the patient were recorded as per the chart review. As part of routine clinical care procedures at our institution, during follow-up visits, neurological side effect subscale scores of the UKU-SERS were recorded (Lingjaerde et al. 1987). The UKU-SERS assesses psychotropic treatment side effects, dividing them into psychic, neurological, autonomic, and other categories. The extent of each side effect is scored between 0 and 3. The scale assesses 48 separate side effects using a semistructured interview. The scale can be used in patients with different psychiatric problems, including psychosis and affective disorders. The scale is an inclusive instrument, with high reliability and validity (Cronbach's α > 0.70) (Lingjaerde et al. 1987). The UKU-SERS comprehensively assesses most antipsychotic-induced side effects. In a systematic review, it was the most commonly used antipsychotic rating scale in 440 studies that examined side effects of antipsychotics (van Strien et al. 2015).

Statistical analysis

The frequency of ADRs, demographics, illness, and treatment characteristics of the patients who developed ADRs, in addition to treatment changes in response to ADRs, were analyzed using descriptive statistics. Univariate comparisons of demographics, type of disorder, side effects, and drug treatment regimens in the patients with and without ADRs were conducted using chi-square, Kruskal–Wallis, or Mann–Whitney U tests, depending on the variable characteristics. The ADR incidence associated with one or two antipsychotics and with specific antipsychotics was compared using a chi-square test. The mean antipsychotic dose and mean chlorpromazine equivalence in patients with or without dystonia were investigated using the Mann–Whitney U test. Finally, all variables significant at the level of p < 0.05 in univariate analyses comparing patients with and without ADRs were entered into a backward elimination, multivariable logistic regression analysis to identify independent risk factors for the development of ADRs. All tests were two-sided, with alpha = 0.05 and with correction for multiple testing.

Results

Sample

Of 5114 children and adolescents referred to the Ankara Yildirim Beyazit University's Department of Child and Adolescent Psychiatry between July 1, 2015 and January 1, 2017, treatment with antipsychotics was initiated in 558 (10.9%). patients. Of these, 117 (20.9%) patients did not have at least two follow-up visits and were thus excluded. Hence, the study sample comprised 441 (8.6%) patients (mean age: 13.3 ± 3.5; range: 4–19 years; males: 5.3%; inpatients: 52.0%) who had commenced treatment with antipsychotics and had at least two follow-up visits. Information on the demographics, type of disorder, side effects, and drug treatment regimen in the total sample are detailed in Table 1. There was no statistical difference in terms of developing ADRs and having comorbid psychiatric disorders (χ² = 0.125; p = 0.72).

Table 1.

Demographic Illness and Treatment Characteristics of Patients With and Without Acute Dystonic Reaction

	Total (n = 441)	Patients without dystonia (n = 411)	Patients with dystonia (n = 30)	t, Z or χ²	p
Age (mean ± SD)	13.3 ± 3.5	13.2 ± 3.5	15.1 ± 2.7	−3.5^a	0.004
Sex, male, n (%)	245 (55.3)	228 (55.4)	17 (56.6)	0.016^b	0.528
Primary diagnosis,^c n (%)
Disruptive, impulse control, and conduct disorders	95 (21.5)	94 (22.8)	1 (3.3)	2.726^b	0.119
ADHD	58 (13.2)	56 (13.6)	2 (15)	1.388^b	0.316
Intellectual disability	57 (12.9)	55 (13.4)	2 (15)	0.098^b	0.490
Autism spectrum disorder	57 (12.9)	54 (13.1)	3 (10)	0.151^b	0.486
Bipolar and related disorders	49 (11.1)	38 (9.2)	11 (36.6)	25.449^b	<0.0001
Depressive disorders	41 (9.3)	39 (9.5)	2 (6.6)	2.121^b	0.199
Anxiety disorders	27 (6.1)	27 (6.5)	0	0.180^b	0.470
Schizophrenia spectrum and other psychotic disorders	23 (5.2)	17 (4.3)	6 (20)	7.768^b	0.015
Substance use disorders	18 (4.1)	15 (3.6)	3 (10)	6.359^b	0.033
Obsessive/compulsive and related disorders	8 (1.8)	8 (1.9)	0	N/A	N/A
Tic disorders	5 (1.1)	5 (1.2)	0 (0)	N/A	N/A
Feeding and eating disorders	3 (0.7)	3 (0.7)	0 (0)	N/A	N/A
Inpatients, n (%)	233 (52.8)	204 (49.6)	29 (96.6)	24.818^b	<0.0001
FGA treatment in single antipsychotic use, n (%)	19 (100)	17 (89.5)	2 (10.5)	4.345^b	0.037
SGA treatment in single antipsychotic use, n (%)	372 (100)	363 (97.6)	9 (2.4)	4.345^b	0.037
FGA treatment (in single or combined antipsychotic use), n (%)	26 (5.9)	18 (4.4)	8 (26.6)	25.032^b	<0.0001
SGA treatment only (in single or combined antipsychotic use), n (%)	415 (94.1)	393 (95.6)	22 (73.3)	25.032^b
Single antipsychotic users, n (%)	391 (88.7)	380 (92.4)	11 (36.6)	86.572^b	<0.0001
Two antipsychotics users, n (%)	50 (11.3)	31 (7.5)	19 (63.3)	86.572^b	<0.0001
Antipsychotic chlorpromazine dose for single antipsychotic users (mean ± SD)	106.5 ± 121.4 (median = 66.6)	104.5 ± 20.2 (median = 66.6)	177.0 ± 146.9 (median = 133.3)	−2.371^d	0.02
Antipsychotic chlorpromazine dose for two antipsychotic users (mean ± SD)	386.9 ± 347.2 (median = 271.6)	285.5 ± 168.1 (median = 235)	552.5 ± 484.2 (median = 300)	−2.931^d	0.003

Student t test was performed.

Chi-square test was performed.

The primary diagnosis of the cases are mentioned, 251 of the cases (56.9%) have comorbid psychiatric diagnosis.

Mann–Whitney U test was performed.

ADHD, attention-deficit/hyperactivity disorder; FGA, first-generation antipsychotic; N/A, not applicable; SD, standard deviation; SGA, second-generation antipsychotic.

The sociodemographic characteristics and dystonia rates of the patients treated with one or two antipsychotic agents are presented in Table 2.

Table 2.

Age and Sex Characteristics of the Patients With and Without Dystonia

	Total	Patients without dystonia	Patients with dystonia	t, Z or χ²	p
Single antipsychotic users (n = 391)
Age (mean ± SD)	13.0 ± 3.5	13.0 ± 3.5	14.4 ± 3.3	− 1.405^a	0.16
Sex, total; male; female, n (%)	391 (100); 221 (56.5); 170 (43.5)	380 (97.2); 216 (56.8); 164 (43.2)	11 (2.8); 5 (45.4); 6 (54.6)	0.564^b	0.45
Two antipsychotics users (n = 50)
Age (mean ± SD)	15.5 ± 2.3	15.5 ± 2.5	15.5 ± 2.2	−0.355^a	0.72
Sex, total; male; female, n (%)	50 (100); 24 (48); 26 (52)	31 (62); 12 (39); 19 (61)	19 (38); 12 (63.1); 7 (36.8)	2.821	0.093

Student t test was performed.

Chi-square test was performed.

SD, standard deviation.

The predominant diagnoses were disruptive, impulse control, and conduct disorders (21.5%), followed by ADHD (13.2%), irritability, and aggression accompanying intellectual impairment (12.9%) (Table 1).

SGAs were the most commonly prescribed antipsychotic agents (94.1%) (Table 1). In terms of individual antipsychotics, risperidone (52.2%) was the most commonly prescribed SGA, followed by aripiprazole (12.2%) and olanzapine (10.4%) (Table 3). The mean antipsychotic treatment duration was 99.5 ± 223.3 days (median: 34 days). The ADR incidence between SGA users in single or combined antipsychotic use was 5.3% (Table 1). Three hundred ninety-one (88.7%) patients were prescribed a single antipsychotic, and 50 (11.3%) patients were prescribed two antipsychotics. The patients prescribed two antipsychotics were significantly older (mean age: 15.5 ± 2.3 years) than those prescribed a single antipsychotic (mean age: 13.0 ± 3.5 years; Z = −4.978; p < 0.0001) (Table 2).

Table 3.

Demographic Treatment Characteristics of Patients Who Developed Dystonia During Use of a Single Antipsychotic

Single antipsychotic users (n = 391, %100)	Risperidone (52.2%)	Aripiprazole (12.2%)	Olanzapine (10.4%)	LAIR (3.8%)	Haloperidol (2.0%)	Zuclopenthixol decanoate (0.6%)
The rate of ADR^a/total patients (11/391), n (%)	3/230 (1.3)	2/54 (3.7)	3/46 (6.5)	1/17 5.9)	1/9 (11.1)	1/3 (33.3)
Age (mean ± SD)	11.0 ± 5.3	15.5 ± 0.7	15.0	17.0	15.0	17.0
Sex, male, n (%)	2 (66.7)	0	2 (66.7)	0	0	1 (100)
Inpatients, n (%)	2 (66.7)	2 (100)	3 (100)	1 (100)	1 (100)	1 (100)
Non-SGA psychotropic comedication, n
Antidepressants	2	1	1	0	0	1
Mood stabilizer	0	1	0	0	1	0
ADHD treatment
Methylphenidate	1	0	1	0	0	0
Atomoxetine	0	2	0		1
Anxiolytics/hypnotics	1	1	0	0	0	0
Time of ADR (days)
	4.6 ± 5.5 (median = 2 days)	4.0 ± 4.2 (median = 4 days)	2.3 ± 1.5 (median = 2 days)	5 days	1 day	1 day
ADR treatment
	Biperiden+antipsychotic change (two patients); only biperiden (one patient)	Biperiden+antipsychotic change	Biperiden+antipsychotic change (one patient); only biperiden (two patients)	Only biperiden	Biperiden+antipsychotic change	Biperiden+antipsychotic change

There were 22 (4.9%) quetiapine, 7 (1.6%) chlorpromazine, 2 (0.4%) paliperidone, 1 (0.2%) clozapine only using patients and no ADR was seen with those.

ADHD, attention-deficit/hyperactivity disorder; ADR, acute dystonic reaction; LAIR, long-acting injectable risperidone; SD, standard deviation; SGA, second-generation antipsychotic.

Four hundred fifteen (94.1%) patients were treated with a single antipsychotic drug or antipsychotic polypharmacy with SGAs, and 26 (5.9%) patients received FGAs (monotherapy or combined therapy).

The incidence of ADRs in patients prescribed FGAs (one or two antipsychotics) was 30.8% (Table 1).

Frequency and characteristics of patients who developed ADRs on antipsychotic monotherapy

Among the 441 patients who commenced treatment with antipsychotics, ADRs developed in 30 (6.8%) patients (males, n = 16/53.3%). ADRs developed in 11 (2.8%) of 391 patients treated with a single antipsychotic and 19 (38.0%) of 50 patients treated with 2 antipsychotics (Table 2).

Among the 11 patients prescribed antipsychotic monotherapy, 5 (45.5%) patients developed ADRs at the initiation dosage (mean time until ADR: 4.0 ± 4.0 days), and the other 6 (54.5%) patients developed ADRs 2.7 ± 2.4 days after increasing the antipsychotic dose (Fig. 1).

Table 2 provides information on the demographic and drug treatment regimens of the patients who developed ADRs during antipsychotic monotherapy. Table 4 provides details on the doses and chlorpromazine equivalences of antipsychotic monotherapy in children and adolescents with and without ADRs. In the study, 33.3%, 11.1%, 6.5%, 5.9%, 3.7%, 1.3%, 0%, and 0% of patients treated with zuclopenthixol, haloperidol, olanzapine, long-acting injectable (LAI) risperidone, aripiprazole, risperidone, quetiapine, and chlorpromazine, respectively, developed ADRs (Table 4). When the data were pooled, 8 of 26 patients treated with FGAs developed ADRs as compared with 22 of 415 patients treated with SGAs. (χ ² = 25.032; p < 0.0001) (Table 1).

Table 4.

Antipsychotic Doses and Chlorpromazine Equivalences in Patients With and Without Acute Dystonic Reaction During Use of a Single Antipsychotic

	Total N = 391	Patients on antipsychotic monotherapy without dystonia (n = 380)			Patients on antipsychotic monotherapy developing dystonia (n = 11)
	n (%)	N (%)	Dose (mg/d)	Chlorpromazine equivalent (mg/d)	N (%)	Dose (mg/d)	Chlorpromazine equivalent (mg/d)
Risperidone	230 (58.8)	227 (98.7)	1.1 ± 0.7	53.3 ± 33.6	3 (1.3)	1.4 ± 1.0	70.8 ± 50.5
Aripiprazole	54 (13.8)	52 (96.3)	7.4 ± 3.8	101.4 ± 52.5	2 (3.7)	7.5 ± 3.5	99.9 ± 47.1
Olanzapine	46 (11.8)	43 (93.5)	10.1 ± 5.4	201.2 ± 107.7	3 (6.5)	13.3 ± 5.8	266.6 ± 115.4
Quetiapine	22 (5.6)	22 (100)	215.9 ± 192.3	287.8 ± 256.4	0	—	—
Haloperidol	9 (2.3)	8 (88.9)	6.8 ± 5.8	341.8 ± 292.0	1 (11.1)	10.0	50
Zuclopenthixol decanoate	3 (0.8)	2 (66.6)	25.0 ± 7.1	166.6 ± 47.2	1 (33.3)	15	100
Long-acting injectable risperidone	17 (4.3)	16 (94.1)	2.7	135	1 (5.9)	2.7	135
Chlorpromazine	7 (1.8)	7 (100)	142.9 ± 60.7		0	—	—
Paliperidone	2 (0.5)	2 (100)	6.0	200	0	—	—
Clozapine	1 (0.3)	1 (100)	350	700	0	0
Total	391	380	N/A	100.9 ± 118.8	11	N/A	181.2 ± 154.1

N/A, not applicable.

Frequency and characteristics of patients who developed ADRs on two antipsychotics

Of 50 patients prescribed 2 antipsychotics, 19 (38%) patients developed ADRs. Of these, 7 (46.8%) patients developed ADRs after the addition of the second antipsychotic (mean time until ADR: 3.0 ± 2.3 days), and 12 (63.2%) patients developed ADRs 1.6 ± 0.8 days after increasing the dosage of the second antipsychotic agent (Fig. 1).

Table 5 provides information on the demographics and drug treatment regimens/protocols of the patients treated with two antipsychotics that developed ADRs and Table 6 presents the doses and chlorpromazine equivalences of each of the two antipsychotics in patients with and without ADRs.

Table 5.

Demographic and Treatment Characteristics of Patients Who Developed Dystonia During Treatment with Two Antipsychotics

	OLZ–ARP (n = 4)	RIS–OLZ (n = 3)	RIS–LAIR (n = 2)	RIS–ARP (n = 1)	OLZ–QUE (n = 1)	LAIR–OLZ (n = 1)	RIS–HLP (n = 1)	OLZ–ZCLP (n = 1)	HLP–CLP (n = 1)	LAIR–ARP (n = 1)	OLZ–CLP (n = 1)	ARP–CLP (n = 1)	RIS–ZCLP (n = 1)
The rate of ADR^a/total patients (19/50)	4/5	3/6	2/5	1/7	1/4	½	1/1	1/1	1/1	1/1	1/1	1/1	1/1
Age (mean ± SD)	15.7 ± 1.2	15.3 ± 2.1	17.5 ± 0.7	14	18	16	16	14	10	13	18	14	16
Sex, male, n (%)	2 (50)	1 (33.3)	1 (50)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	0	0	1 (100)	1 (100)	1 (100)
Inpatients, n (%)	4 (100)	3 (100)	2 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)	1 (100)
Non-SGA psychotropic comedication, n (%)
Antidepressants	2 (50)	3 (100)	1 (50)	0	0	0	0	0	0	1 (100)	0	1 (100)	0
Mood stabilizer	3 (75)	1 (33.3)	1 (50)	1 (100)	1 (100)	1 (100)	1 (100)	0	1 (100)	0	1 (100)	1 (100)	1 (100)
ADHD treatment
Methylphenidate	0	0	0	0	0	0	0	0	0	0	0	0	0
Atomoxetine	0	0	0	0	0	0	0	0	0	0	0	0	0
Anxiolytics/hypnotic	1 (25)	1 (33.3)	1 (50)	0		0	0	1 (100)	0	0	1 (100)	0	0
Time of ADR (days)
	1.7 ± 0.9 (median = 1.5 days)	1.3 ± 0.5 (median = 1 day)	5.5 ± 3.5 (median = 5.5 days)	2 days	2 days	3 days	1 day	1 day	1 day	2 days	2 days	1 day	3 days
ADR treatment
	Biperiden+antipsychotic change (1 patient); biperiden+antipsychotic dose was decreased (3 patients)	Biperiden+antipsychotic change (1 patient); biperiden+antipsychotic dose was decreased (1 patients); only biperiden (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient); only biperiden (1 patient)	Biperiden+antipsychotic change (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient)	Only biperiden (1 patient)	Only biperiden (1 patient)	Only biperiden (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient)	Biperiden+antipsychotic dose was decreased (1 patient)	Biperiden+antipsychotic change (1 patient)

There were 12 RIS–QUE, 1 ARP–QUE, 1 OLZ–ZCLP combination using patients that did not develop ADR.

ADHD, attention-deficit/hyperactivity disorder; ADR, acute dystonic reaction; ARP, aripiprazole; CLP, chlorpromazine; HLP, haloperidol; LAIR, long-acting injectable risperidone; OLZ, olanzapine; QUE, quetiapine; RIS, risperidone; SD, standard deviation; SGA, second-generation antipsychotic; ZCLP, zuclopenthixol decanoate.

Table 6.

Antipsychotic Doses and Chlorpromazine Equivalences in Patients With and Without Acute Dystonic Reaction During Use of Two Antipsychotics

	Total (n = 50)	Patients on two antipsychotics without dystonia (n = 31)							Patients on two antipsychotics with dystonia (n = 19)
		N (%)	Mean dose of first antipsychotic (mg/d)	Chlorpromazine equivalent first antipsychotic (mg/d)	Mean dose of second antipsychotic (mg/d)	Chlorpromazine equivalent second antipsychotic (mg/d)	Total chlorpromazine equivalent (mg/d)	N (%)	Mean dose of first antipsychotic (mg/d)	Chlorpromazine equivalent first antipsychotic (mg/d)	Mean dose of second antipsychotic (mg/d)	Chlorpromazine equivalent second antipsychotic (mg/d)	Total chlorpromazine equivalent (mg/d)
Risperidone-quetiapine	12	12 (100%	2.0 ± 0.9	97.9 ± 45.7	139.6 ± 87.6	186.1 ± 116.7	284.0 ± 125.	—	—	—	—	—	—
Risperidone-aripiprazole	7	6 (85.7)	1.3 ± 0.4	62.5 ± 20.9	7.3 ± 2.9	97.6 ± 39.2	160.2 ± 55.1	1 (14.3)	2.0	100	15.0	200	300
Risperidone-olanzapine	6	3 (50)	1.5 ± 0.5	75.0 ± 25.0	6.7 ± 2.9	133.3 ± 57.7	208.3 ± 80.4	3 (50)	1.3 ± 0.6	66.6 ± 28.8	12.50 ± 2.50	250 ± 50	316.6 ± 76.4
Risperidone-LAIR	5	3 (60)	2.3 ± 0.6	116.6 ± 28.8	25.0	135	251.6 ± 28.8	2 (40)	2.0 ± 1.4	100 ± 70.7	25.0	135	235.0 ± 70.7
Olanzapine-quetiapine	4	3 (75)	15.0 ± 8.7	300 ± 173.2	200.0 ± 100	266.6 ± 133.3	596.6 ± 296.3	1 (25)	30	600	900	1200	1800
Olanzapine-aripiprazole	5	1 (20)	20	400	15	200	600	4 (80)	18.75 ± 10.3	375 ± 206.1	11.2 ± 4.7	150 ± 63.8	524.9 ± 185.3
LAIR-olanzapine	2	1 (50)	5.0	100	25.0	135	235	1 (50)	30.0	600	25.0	135	735
Risperidone-haloperidol	1	0	—	—	—	—	—	1 (100)	2.0	100	7.5	375	475
Aripiprazole-quetiapine	1	1 (100)	7.5	100	200.0	266	366	0	—	—	—	—	—
Olanzapine-zuclopenthixol acetate	2	1 (50)	5.0	100	15	100	200	1 (50)	5	100	15	100	200
Haloperidol-chlorpromazine	1	0	—	—	—	—	—	1 (100)	2.0	100	300.0	300.0	400
LAIR-aripiprazole	1	0	—	—	—	—	—	1 (100)	2.7	135	10.0	133.3	268.3
Olanzapine-chlorpromazine	1	0	—	—	—	—	—	1 (100)	30.0	600	300.0	300.0	900
Aripiprazole-chlorpromazine	1	0	—	—	—	—	—	1 (100)	15.0	200	300.0	300.0	500
Risperidone-zuclopenthixol acetate	1	1	2	100	15	100	200	0	—	—	—	—	—
Total	50	31	N/A	120.1 ± 99.4	N/A	172.5 ± 105.1	285.5 ± 168.1	19	N/A	312.3 ± 357.8	N/A	282.7 ± 277.9	526.6 ± 485.1

LAIR, long-acting injectable risperidone; N/A, not applicable.

Treatment changes in response to ADRs

In 12 (40.0%) of 30 ADR cases, the antipsychotic deemed to be responsible for the ADR was discontinued. Each of the remaining 18 cases received the anticholinergic, biperiden, either as the sole treatment (n = 10, 33.3%) or in addition to lowering the dose of the antipsychotic believed to be responsible for the ADR (n = 8, 26.6%) (Fig. 1). The mean biperiden continuation times during a follow-up of 6 months are presented in Figure 1.

Multivariable correlates of ADRs

The following factors were significantly different (p < 0.05) in patients with and without ADRs: older age; inpatient status; multiple antipsychotic use; higher chlorpromazine equivalence antipsychotic doses; FGA use; diagnoses of bipolar disorder, schizophrenia, or substance use disorders; and mood stabilizers (Table 1). All the significant variables in the analyses were entered into a multivariable logistic regression model. After performing backward elimination logistic regression analysis, the remaining significant and independent correlates of ADRs were as follows: cotreatment with two antipsychotics (odds ratio [OR]: 14.255; p < 0.001), inpatient treatment (OR: 13.866; p = 0.013), FGA use (OR: 4.730; p = 0.015), a diagnosis of schizophrenia (OR: 3.716; p = 0.039), and a diagnosis of bipolar disorder (OR: 2.898; p = 0.046; r² = 0.451, p < 0.0001) (Table 7).

Table 7.

Independent Correlates of Acute Dystonic Reactions During Antipsychotic Treatment

Variable	B	OR	CI	p
Multiple antipsychotic use	2.657	14.255	5.659–35.912	<0.0001
Inpatient treatment	2.629	13.866	1.748–110.018	0.013
Using FGAs	1.554	4.730	1.356–16.504	0.015
Being diagnosed with schizophrenia	1.313	3.716	1.067–12.943	0.039
Being diagnosed with bipolar disorder	1.064	2.898	1.017–8.259	0.046

p < 0.05 accepted as significant. Overall model: r ² = 0.451, p < 0.0001.

CI, confidence interval; FGAs, first-generation antipsychotics; OR, odds ratio.

Discussion

In this study, the incidence of ADRs in 441 children and adolescents following the initiation of antipsychotics was 6.8% overall, with ADRs developing within 2–4 days after commencement of treatment with a new antipsychotic or an increase in the drug dose. An important finding was that the rate of ADRs in patients treated with a single FGA was higher than that in patients treated with a single SGA (10.5% vs. 2.4%). The risk of ADRs varied from low risk with the low-potency FGA chlorpromazine to mid-/high-potency FGAs, zuclopenthixol, haloperidol, and from the quetiapine, paliperidone, oral risperidone, and aripiprazole to risperidone LAI and olanzapine among the SGAs. Independent correlates of ADRs included cotreatment with two antipsychotics, inpatient treatment, FGA use, and diagnoses of schizophrenia or bipolar disorder.

In the present study, the ADR incidence in children and adolescents following the commencement of antipsychotic treatment was 6.8%. Among those taking a single antipsychotic agent, the ADR incidence was 2.8%, whereas it was 38% in those who received cotreatment with two antipsychotics. In a previous study, the incidence of ADRs in 1152 psychiatric inpatients who received phenothiazine, butyrophenone, or thioxanthene was 10.1% (Swett 1975). In a study that examined ADRs in first-episode psychotic patients, 60% of patients treated with haloperidol alone developed dystonia (Aguilar et al. 1994). Although precise estimates of the incidence of dystonic reactions are not available, such reactions appear to be less common among patients using low-potency FGAs and relatively rare among those using SGAs (American Psychiatric Association 2004). In a study on 1337 adults treated with antipsychotics, the ADR incidence was 3.1%. Among 41 patients who developed ADRs, 32 patients used FGAs, and four of the risperidone-using cases developed ADRs after emergency parenteral treatment with typical neuroleptics (Raja and Azzoni 2001). In a retrospective study, the incidence of neuroleptic-induced dystonia was 15.7% (Ballerini et al. 2002). The prevalence of ADRs in patients treated with conventional neuroleptics has been estimated to range from 2.3% to 60% (Cunningham Owens 2014) as compared with 2% to 3% in patients treated with atypical neuroleptics (Pierre 2005).

Another important finding of the present study was the speed of onset of ADRs, with ADRs developing 2–4 days after starting a new antipsychotic or increasing the drug dose. Dystonic reactions frequently arise after the first few doses of medication, and 90% occur within the first 3 days (American Psychiatric Association 2004). Thus, ADRs respond dramatically to the administration of anticholinergic or antihistaminic medication.

In the present study, the ADR incidence in FGA-treated patients was higher (30.8%) than that in SGA-treated cases (5.3%; single or combined use). However, no ADRs occurred in the patients treated with chlorpromazine only. This finding may be due to the low potency of chlorpromazine and its higher affinity for histaminergic receptors (Correll 2010). In addition, no ADRs developed in the patients treated with quetiapine or clozapine alone. As reported previously, these antipsychotics have a stronger affinity for histaminergic and cholinergic receptors than dopaminergic receptors and therefore are associated with fewer ADRs (Correll 2010). In the present study, two patients were treated with paliperidone, neither of whom developed ADRs. As is well known, the time to the peak concentration and half-life of antipsychotics can help to predict the speed of onset of side effects (Correll 2010). Previous studies reported that the time to the peak concentration of paliperidone was as long as 24 hours, whereas that of risperidone was 1–2 hours (Correll 2010; De Leon et al. 2010). Despite the difference in the time to the peak concentration of paliperidone, we could not decide with two cases. Further studies are needed to understand this difference.

In our study, the frequency of ADRs in patients treated with LAI risperidone appeared to be higher than that in patients treated with oral risperidone. In our previous study in a naturalistic setting, 16.7% of patients treated with 25 mg of LAI risperidone reported muscle spasms, with the patients subjectively reporting that muscle spasms were one of the most common side effects (Ceylan et al. 2017). However, a review reported no significant difference between injectable and oral forms of antipsychotics in terms of EPS side effects (Misawa et al. 2016). Further studies are needed to determine the potential ADR risk associated with LAI risperidone.

One of three patients treated with zuclopenthixol decanoate developed an ADR in the present study. Data on the incidence of EPS side effects with depot formulations are unclear. Some studies suggested that the incidence of EPS side effects was significantly higher in patients receiving depot formulations, whereas other studies found no difference in the incidence of side effects between oral and depot antipsychotics (Altamura et al. 2003). Studies on the incidence of ADRs in patients treated with zuclopenthixol decanoate would be useful.

Another finding of this study is the correlates of ADRs in children and adolescents. Cotreatment with two antipsychotics, inpatient treatment, FGA use, and diagnoses of schizophrenia or bipolar disorder are the predictors of ADR. In literature, in addition to the use of high-potency medications, other risk factors for dystonic reactions are young age, male sex, high doses, and intramuscular administration (American Psychiatric Association 2004). There was no difference between cases with and without ADR in terms of age and sex in this study. This may be due to the age of the participants. In their study, Raja and Azzoni (2001) also did not find a difference between sexes but the ADR cases were significantly younger (33.7 ± 11.6 vs. 42.0 ± 14.1 years). Their cases were adults. Further studies are needed to understand the effect of age on the development of ADRs in children and adolescents.

This study revealed a high risk of ADRs among inpatients. Psychiatric conditions requiring hospitalization are usually severe or resistant to treatment (Bardach et al. 2014). Inpatients may have various severe symptoms, such as severe aggression, manic episodes, psychotic symptoms, and suicidal tendencies, which are hard to bear for family and/or institutions. During the hospitalization period, aggressive dosing, high-than-normal doses, and antipsychotic polypharmacy may be needed to control the symptoms. A review study by Kumar et al. (2013) also showed that antipsychotics can cause dystonia at low doses, however the risk increases as the dose increases. In our study, the use of combined antipsychotics was one of the strongest predictors of the development of ADRs. Treatment with combined antipsychotics may increase the risk of ADRs by blocking dopaminergic receptors in various ways. Thus, the findings point to the need for being careful when using multiple antipsychotics in children and adolescents in terms of ADRs. The incidence of bipolar disorder and schizophrenia cases among inpatients is higher than that among outpatients. Thus, inpatient treatment causes high dose and/or multiple drug combinations. That could explain why the inpatient treatment is a risk factor for ADRs.

Antipsychotic polypharmacy is another predictor of ADRs. Previous studies reported that a diagnosis of schizophrenia, severe long-lasting symptoms, and hospitalization rates are potential reasons for antipsychotic polypharmacy (Gilmer et al. 2007; Correll and Gallego 2012; Sneider et al. 2015). In a study on adults, antipsychotic polypharmacy was associated with greater anticholinergic requirements; shorter observation times, greater illness severity, and lower antidepressant use (Gallego et al. 2012). Antipsychotic polypharmacy was also reported in child and adolescent populations, with a mean prevalence of 9.6 ± 7.2 in a systematic review (Toteja et al. 2014). In the present study, antipsychotic polypharmacy was related to an increased incidence of ADRs. This may be due to higher total antipsychotic doses as chlorpromazine equivalents. Although a higher CPZ dose was significant in the univariate analyses, it was not significant in the logistic regression model. Thus, the increased incidence of ADRs with antipsychotic polypharmacy may be related to drug–drug interactions and synergic effects of the antipsychotics (Stahl 2015).

Limitations and strengths of the study

The first limitation is that this is a chart review study and not a prospective study, and the antipsychotic treatments are not randomized. Therefore, only associations are reported, and causality cannot be determined. However, this naturalistic feature also enhances the generalizability of the findings to other usual care settings. The second limitation is the limited set of predictors. The predictors are restricted to clinical variables, and ADR is the only EPS side effect assessed. The third limitation is the absence of drug blood level measurements. Despite these limitations, to our knowledge, this is the largest study on ADR side effects of antipsychotics in children and adolescents. The treatment of the consecutive cohort according to usual care procedures is an additional strength of the study. The routine UKU-SERS assessments, CPZ dose information, and multivariable analyses also strengthen the study. Finally, the assessments of the side effects (ADRs) of single antipsychotics and combined antipsychotic treatment are strengths of this study.

Conclusions

In conclusion, in terms of drug safety, atypical antipsychotics do not seem to be associated with a high risk of ADRs in children and adolescents. Mid-/high-potency FGAs are more likely to cause ADRs in children and adolescents. ADRs, which are rare with single antipsychotic usage in children and adolescents, are relatively common with combined antipsychotics. Despite its rare side effect, it should be handled as a serious side effect that may require drug changes and therefore limit treatment options. Depot formulations of antipsychotics need to be screened in terms of ADR development. The effect of sex and age on the development of ADRs in children and adolescents seems different than in adults. Further studies are needed to understand their effects. Finally, the independent correlates of ADRs included cotreatment with two antipsychotics, inpatient treatment, FGA use, and diagnoses of schizophrenia or bipolar disorder. Additional studies are needed to understand the impact of drug–drug interactions on the incidence of ADRs.

Clinical Significance

These findings provide a frequency estimate about ADRs in different situations like using a single or double antipsychotic, FGA or SGA, and may affect the antipsychotic selection in children and adolescents. The other important clinical finding is that the antipsychotic polypharmacy, inpatient treatment, FGA use, and diagnoses of schizophrenia or bipolar disorder are independent factors associated with ADRs.

Footnotes

Acknowledgments

The authors would like to thank the adolescent patients and their families who participated in the study.

Disclosures

Dr. Correll has been a consultant and/or advisor to or has received honoraria from: Alkermes, Allergan, Angelini, Gerson Lehrman Group, IntraCellular Therapies, Janssen/J&J, LB Pharma, Lundbeck, Medavante, Medscape, Merck, Neurocrine, Otsuka, Pfizer, ROVI, Servier, Sunovion, Takeda, and Teva. He has provided expert testimony for Bristol-Myers Squibb, Janssen, and Otsuka. He served on a Data Safety-Monitoring Board for Lundbeck and ROVI. He received royalties from UpToDate and grant support from Janssen and Takeda. He is also a shareholder of LB Pharma. The other authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Abdo

, van de Warrenburg

, Burn

, Quinn

, Bloem

: The clinical approach to movement disorders. Nat Rev Neurol, 6:29–37, 2010.

Aguilar

, Keshavan

, Martínez-Quiles

, Hernández

, Gómez-Beneyto

, Schooler

: Predictors of acute dystonia in first-episode psychotic patients. Am J Psychiatry, 151:1819–1821, 1994.

Alacqua

, Trifirò

, Arcoraci

, Germanò

, Magazù

, Calarese

, Di Vita

, Gagliano

, Spina

: Use and tolerability of newer antipsychotics and antidepressants: A chart review in a paediatric setting. Pharm World Sci, 30:44–50, 2008.

Altamura

, Sassella

, Santini

, Montresor

, Fumagalli

, Mundo

: Intramuscular preparations of antipsychotics: Uses and relevance in clinical practice. Drugs, 63:493–512, 2003.

American Psychiatric Association: Practice guideline for the treatment of patients with schizophrenia, second edition. Am J Psychiatr 161(2, Suppl):1–56, 2004.

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 5th ed. Arlington, VA: American Psychiatric Association; 2013.

Ballerini

, Bellini

, Niccolai

, Pieroni

, Ferrara

: Neuroleptic-induced dystonia: Incidence and risk factors. Eur Psychiatry, 17:366–368, 2002.

Bardach

, Coker

, Zima

, Murphy

, Knapp

, Richardson

, Edwall

, Mangione-Smith

: Common and costly hospitalizations for pediatric mental health disorders. Pediatrics, 133:602–609, 2014.

Carbon

, Kapoor

, Sheridan

, Al-Jadiri

, Azzo

, Sarkaria

, Kane

, Saito

, Correll

: Neuromotor adverse effects in 342 youth during 12 weeks of naturalistic treatment with five second-generation antipsychotics. J Am Acad Child Adolesc Psychiatry, 54:718–727, 2015.

10.

Ceylan

, Erdogan

, Tural Hesapcioglu

, Cop

: Effectiveness, adverse effects and drug compliance of long-acting injectable risperidone in children and adolescents. Clin Drug Investig, 37:947–956, 2017.

11.

Christian

, Saavedra

, Gaynes

, Sheitman

, Wines

RCM

, Jonas

, Viswanathan

, Ellis

, Woodell

, Carey

: Future Research Needs for First- and Second-Generation Antipsychotics for Children and Young

Adults

. AHRQ Future Research Needs Papers, No. 13. Rockville, MD: Agency for Healthcare Research and Quality (US); February 2012.

12.

Correll

: Antipsychotic use in children and adolescents: Minimizing adverse effects to maximize outcomes. J Am Acad Child Adolesc Psychiatry, 47:9–20, 2008.

13.

Correll

: From receptor pharmacology to improved outcomes: Individualizing the selection, dosing, and switching of antipsychotics. Eur Psychiatry 25(Suppl, 2):S12–S21, 2010.

14.

Correll

, Gallego

: Antipsychotic polypharmacy: A comprehensive evaluation of relevant correlates of a long-standing clinical practice. Psychiatr Clin North Am, 35:661–681, 2012.

15.

Correll

, Penzner

, Parikh

, Mughal

, Javed

, Carbon

, Malhotra

: Recognizing and monitoring adverse events of second-generation antipsychotics in children and adolescents. Child Adolesc Psychiatr Clin N Am, 15:177–206, 2006.

16.

Cossu

, Colosimo

: Hyperkinetic movement disorder emergencies. Curr Neurol Neurosci Rep, 17:6, 2017.

17.

Cunningham Owens

: A Guide to the Extrapyramidal Side-Effects of Antipsychotic Drugs, 2nd ed. Cambridge,, England: Cambridge Press, 2014.

18.

De Leon

, Wynn

, Sandson

: The pharmacokinetics of paliperidone versus risperidone. Psychosomatics, 51:80–88, 2010.

19.

Derinoz

, Caglar

: Drug-induced movement disorders in children at paediatric emergency department: “Dystonia.”. Emerg Med J, 30:130–133, 2013.

20.

Gallego

, Bonetti

, Zhang

, Kane

, Correll

: Prevalence and correlates of antipsychotic polypharmacy: A systematic review and meta-regression of global and regional trends from the 1970s to 2009. Schizophr Res, 138:18–28, 2012.

21.

Gilmer

, Dolder

, Folsom

, Mastin

, Jeste

: Antipsychotic polypharmacy trends among Medicaid beneficiaries with schizophrenia in San Diego County, 1999–2004. Psychiatr Serv, 58:1007–1010, 2007.

22.

Kumar

, Datta

, Wright

, Furtado

, Russell

: Atypical antipsychotics for psychosis in adolescents. Cochrane Database Syst Rev, 15:CD009582, 2013.

23.

Leucht

, Cipriani

, Spineli

, Mavridis

, Orey

, Richter

, Samara

, Barbui

, Engel

, Geddes

, Kissling

, Stapf

, Lässig

, Salanti

, Davis

: Comparative efficacy and tolerability of 15 antipsychotic drugs inschizophrenia: A multiple-treatments meta-analysis. Lancet, 382:951–962, 2013.

24.

Lingjaerde

, Ahlfors

, Bech

, Dencker

, Elgen

: The UKU side effect rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr Scand Suppl, 334:1–100, 1987.

25.

Menard

, Thümmler

, Auby

, Askenazy

: Preliminary and ongoing French multicenter prospective naturalistic study of adverse events of antipsychotic treatment in naive children and adolescents. Child Adolesc Psychiatry Ment Health, 13:18, 2014.

26.

Misawa

, Kishimoto

, Hagi

, Kane

, Correll

: Safety and tolerability of long-acting injectable versus oral antipsychotics: A meta-analysis of randomized controlled studies comparing the same antipsychotics. Schizophr Res, 176:220–230, 2016.

27.

Olfson

, Blanco

, Liu

, Wang

, Correll

: National trends in the office-based treatment of children and adolescents, and adults with antipsychotics. Arch Gen Psychiatry, 69:1247–1256, 2012.

28.

Olfson

, Blanco

, Wang

, Laje

, Correll

: National trends in the mental health care of children and adolescents by office-based physicians. JAMA Psychiatry, 71:81–90, 2014.

29.

Penfold

, Stewart

, Hunkeler

, Madden

, Cummings

, Owen-Smith

, Rossom

, Lu

, Lynch

, Waitzfelder

, Coleman

, Ahmedani

, Beck

, Zeber

, Simon

. Use of antipsychotic medications in pediatric populations: What do the data say?. Curr Psychiatry Rep, 15:426, 2013.

30.

Pierre

: Extrapyramidal symptoms with atypical antipsychotics: Incidence, prevention and management. Drug Saf, 28:191–208, 2005.

31.

Pringsheim

, Doja

, Belanger

, Patten S; Canadian Alliance for Monitoring Effectiveness and Safety of Antipsychotics in Children (CAMESA) Guideline

Group

: Treatment recommendations for extrapyramidal side effects associated with second-generation antipsychotic use in children and youth. Paediatr Child Health, 16:590–598, 2011.

32.

Raja

, Azzoni

: Novel antipsychotics and acute dystonic reactions. Int J Neuropsychopharmacol, 4:393–397, 2001.

33.

Sarkar

, Grover

: Antipsychotics in children and adolescents with schizophrenia: A systematic review and meta-analysis. Indian J Pharmacol, 45:439–446, 2013.

34.

Satterwaite

, Wolf

, Rosenheck

, Gur

, Caroff

: A meta-analysis of the risk of acute extrapyramidal symptoms with intramuscular antipsychotics for the treatment of agitation. J Clin Psychiatry, 69:1869–1879, 2008.

35.

Seida

, Schouten

, Mousavi

, Hamm

, Beaith

, Vandermeer

, Dryden

, Boylan

, Newton

, Carrey

: First- and Second Generation Antipsychotics for Children and Young Adults [Internet]. Rockville, MD: Agency for Healthcare Research and Quality (US); 2012, Report No. 11-EHC077-EF.

36.

Sikich

, Frazier

, McClellan

, Findling

, Vitiello

, Ritz

, Ambler

, Puglia

, Maloney

, Michael

, De Jong

, Slifka

, Noyes

, Hlastala

, Pierson

, McNamara

, Delporto-Bedoya

, Anderson

, Hamer

, Lieberman

: Double-blind comparison of first- and second-generation antipsychotics in early-onset schizophrenia and schizo-affective disorder: Findings from the treatment of early-onset schizophrenia spectrum disorders (TEOSS) study. Am J Psychiatry, 165:1420–1431, 2008.

37.

Sikich

, Hamer

, Bashford

, Sheitman

, Lieberman

: A pilot study of risperidone, olanzapine, and haloperidol in psychotic youth: A double-blind, randomized, 8-week trial. Neuropsychopharmacology, 29:133–145, 2004.

38.

Sneider

, Pristed

, Correll

, Nielsen

: Frequency and correlates of antipsychotic polypharmacy among patients with schizophrenia in Denmark: A nation-wide pharmacoepidemiological study. Eur Neuropsychopharmacol, 25:1669–1676, 2015.

39.

Stahl

: Stahl's Essential

Psychopharmacology

, 1st ed. Cambridge University; 2015, pp.129–236.

40.

Swett

, Jr: Drug-induced dystonia. Am J Psychiatry, 132:532–534, 1975.

41.

Toteja

, Gallego

, Saito

, Gerhard

, Winterstein

, Olfson

, Correll

: Prevalence and predictors of antipsychotic polypharmacy in children and adolescents. Int J Neuropsychopharmacol, 17:1095–1105, 2014.

42.

van Strien

, Keijsers

, Derijks

, van Marum

. Rating scales to measure side effects of antipsychotic medication: A systematic review. J Psychopharmacol, 29:857–866, 2015.