Atypical Antipsychotic Medications to Control Symptoms of Delirium in Children and Adolescents

Design

Retrospective chart review

N= 110

Objective

To describe the use of atypical antipsychotics in controlling symptoms of delirium in children and adolescents

Study Groups

Olanzapine (n=78)

Risperidone (n=13)

Quetiapine (n=19)

Inclusion Criteria

Aged 1 to 18 years; admitted to single children's hospital; had a diagnosis code for delirium

Exclusion Criteria

None described

Methods

Pharmacy records at Children’s Hospital Los Angeles were reviewed to identify children who were given antipsychotic agents for delirium. During the period reviewed, the clinical diagnosis of delirium was based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria in association with psychiatric consult.

At the time, oral formulations of olanzapine, risperidone, and quetiapine, and oral and parenteral haloperidol were available on the hospital’s formulary. The oral medications were administered by mouth or via nasogastric, gastric, or gastrojejunal tube. Dosing was estimated on age and weight, and the antipsychotic medications were typically administered at bedtime. 

During this chart review, Delirium Rating Scale-Revised-98 (DRS-R98) scores were retrospectively calculated when possible at time antipsychotic was started to confirm the initial diagnosis of delirium and evaluate symptom severity, and again when antipsychotic was stopped, to assess symptom response.

Duration

January 1, 2005 to December 31, 2006

Outcome Measures

Delirium Rating Scale-Revised-98 (DRS-R98) scores

Baseline Characteristics

Olanzapine (n=78)

Risperidone (n=13)

p-value

Age, years

Mean ± SD

Median (range)

10.8 ± 4.9

12 (1–18)

8.6 ± 5.4

8 (1–16)

10.8 ± 4.5

11 (3–18)

Average daily dose, mg (range)

10 (1–52.5)

1.3 (0.375–4)

56 (12.5–125)

Days on medication (range)

26.5 (1–132)

17.5 (2–54)

35.1 (1–108)

Results

p-value

Pretreatment DRS-R98 score

Mean ± SD

Median (range)

20.4 ± 5.0

20 (11–32)

21.0 ± 5.7

22 (11–30)

16.9 ± 5.1

18 (8–32)

Posttreatment DRS-R98 score

Mean ± SD

Median (range)

4.7 ± 3.0

4 (1–13)

6.7 ± 4.3

6 (2–13)

4.3 ± 3.0

4 (1–11)

Pre-post score difference

Mean ± SD

Median (range)

-15.7 ± 5.6

-15 (4–29) 

-15.3 ± 6.0

-14 (6–24)

-12.4 ± 5.2

-12 (3–25)

Adverse Events

One patient on olanzapine developed mild dystonia which resolved when the dose decreased. No other significant adverse effects were noted.

No definitive statement comparing the efficacy or safety of the three antipsychotic agents can be made with this study design.

Study Author Conclusions

Although randomized placebo-controlled studies are needed, atypical antipsychotic medications appeared to be effective and safe for managing delirium symptoms in pediatric patients while underlying etiology was addressed.

InpharmD Researcher Critique

This was a single-center, retrospective study with no randomization, the medication group sizes were unequal, no placebo/control group was used, DRS-R98 scoring was not blinded, and drug choice and baseline delirium severity were not without potential bias.

This retrospective descriptive study included children as young as 1 year, who have a limited sense of time or place. Additionally, the DRS-R98 score could not be completely scored when the patient was asleep, nonverbal, or sometimes when intubated. Because this study was retrospective, it relied on accurate documentation by providers at the time. Any errors or omissions there may have impacted the DRS-R98 score that was retrospectively calculated. Additionally, the DRS-R98 has not been evaluated for use in pediatric patients.

Olanzapine reduces delirium symptoms in the critically ill pediatric patient

Design

Retrospective, longitudinal study 

N= 59

Objective

To evaluate the efficacy of olanzapine for the resolution of delirium symptoms in the pediatric intensive care unit (PICU) population

Study Groups

Olanzapine (n=31)

Control, no antipsychotic medication (n=28)

Inclusion Criteria

Clinically diagnosed with delirium receiving olanzapine for symptom management of delirium; did not receive any antipsychotic medication for the treatment of delirium 

Exclusion Criteria

None reported

Methods

This was a retrospective study of a Children's Hospital PICU or cardiothoracic intensive care unit (CTICU) from an institution in California. Children diagnosed with delirium were included if they were given olanzapine or no antipsychotic for the treatment of delirium (for various reasons, most commonly due to parental choice). Olanzapine was the only antipsychotic administered in this study.

The delirium rating scale (DRS) was done retrospectively via chart review. This assessment was done to rate delirium severity at the time of receive evaluation. 

When estimating the starting dose of olanzapine, clinicians used patient age, size, and recent use of sedatives or opioids. Common starting doses were 0.625 mg PO QHS to BID for infants and 1.25 mg PO QHS to BID for toddlers. If the patient was older, larger, or experienced more agitation, a scheduled dose starting at 2.5 mg to 5 mg PO QHS to BID was used.

If the patient experienced adverse drug events (ADEs) such as hallucinations, confusion, or insomnia, additional doses of olanzapine would be used PRN; PRN olanzapine dosing ranged from half to equal the starting dose every hour, with a max of four doses/24 hrs. Routine daily dose of olanzapine was also established by the amount of olanzapine required over a 24-hour period. Administration of olanzapine varied from oral, sublingual, nasogastric, or gastronomy tube only. 

To examine a relationship between age and change in delirium severity, age was recorded in five blocks: 0-1 years old, 2-5 years old, 6-10 years old, 11-15 years old, and 16-20 years old.

Initial delirium severity was recorded based on initial DRS in six blocks: 0-5, 6-10, 11-15, 16-20, 21-25, and 26-30.

Duration

January 24, 2006 to July 24, 2010

Outcome Measures

Baseline Characteristics

Olanzapine (n=31)

Control (n=28)

Age, years

Female

Initial DRS score

Results

Olanzapine (n=31)

Control (n=28)

p-value

9.74 ± 7.53

Analysis of variance (ANOVA) results based on unadjusted data revealed greater improvement of delirium in the olanzapine group than in the control.

Adverse Events

No adverse events were associated with olanzapine administration

Study Author Conclusions

This study demonstrates that olanzapine administration is associated with a reduction in the severity of delirium symptoms in pediatric ICU and CTICU patients. Given the findings of this study, olanzapine administration may be a useful therapy for critically ill pediatric patients diagnosed with delirium.

InpharmD Researcher Critique

This study was not randomized and was retrospective which brings to question the establishment of a relationship seen. The retrospective design relies on proper documentation of healthcare providers, which some may have omitted certain information. This is important because delirium rating scale (DRS) scores were retrospectively assessed based on chart documentation. The retrospective use of the DRS used has been studied previously, but its validity has not been established in pediatric patients. 

Additionally, concurrent administration of other medications, such as benzodiazepines and opiates, were not investigated, which may have led to symptoms of delirium. Most importantly, the total dosage for each patient was not available, which does not allow a correlation between olanzapine dose and improvement in delirium symptoms. A robust exclusion or inclusion criteria were not mentioned.

Design

Retrospective, matched, cohort study 

N= 30

Objective

To assess pharmacological treatment in critically ill children treated in respect to a delirium protocol to compare antipsychotic treated patients to non-pharmacologically treated patients 

Study Groups

Untreated (n=15)

Treated (n=15)

Inclusion Criteria

Children treated with antipsychotics for delirium

Exclusion Criteria

Patients with missing data

Methods

This was a retrospective review of a single pediatric intensive care unit (PICU) in Maryland. Antipsychotics available during the review period were haloperidol, risperidone, and quetiapine. If the patient had hypoactive delirium, risperidone would be given for patients <10 years of age and quetiapine if patients were ≥ 10 years of age. Haloperidol was the agent of choice for hyperactive delirium, those unable to take enteral medications, and for patients who required IV use. Pharmacologic therapies were administered as monotherapy and other medications could not be used as prophylaxis for extrapyramidal symptoms (EPS). 

Baseline electrocardiogram (ECG) was retrieved before initiation of pharmacologic therapy. Daily ECG was obtained until adequate dose of antipsychotic drug was established or if medications that could increase QTc was added. Delirium was assessed using the Cornell Assessment of Pediatric Delirium (CAPD) screening tool (score ≥9 indicated a positive delirium screening); diagnoses were confirmed by a PICU physician.

Children treated with antipsychotics for delirium were matched with similar patients who had similar CAPD scores but did not receive pharmacologic treatment for delirium.

Sedation protocol for mechanically ventilated patients involved low-dose fentanyl or dexmedetomidine infusion for short-term sedation (≤48 hours), and for long-term sedation (>48 hours), fentanyl infusion and intermittent low-dose lorazepam was used.

Duration

December 1, 2013 to September 30, 2015

Outcome Measures

Primary endpoints: utilization of sedation medications before and after positive delirium screening and utilization of sedation medications versus non-treated group (no antipsychotic use)

Baseline Characteristics

Untreated (n=15)

Treated (n=15)

Age, years (IQR)

Female

Weight, kg (IQR)

Prior psychiatric history

2 (13.3%)

2 (13.3%)

CAPD delirium type

Hyperactive

Hypoactive

Mixed

0 (0%)

5 (33.3%)

10 (66.7%)

3 (20%)

3 (13.3%)

10 (66.7%)

Include relevant baseline characteristics that will provide a general (big picture) view of the patients in the study.

Results

Untreated (n=15)

Days of delirium (IQR)

Length of mechanical ventilation (IQR)

Length of PICU stay, days (IQR)

Disposition

Discharged home

Rehab

Hospital transfer

Died

14 (93.3%)

1 (7.1%)

0

0

7 (46.7%)

5 (33.3%)

1 (6.7%)

2 (13.3%)

Haloperidol, risperidone, and quetiapine were used in 9, 6, and 2 patients, respectively. Ten patients in the treatment group had improved delirium scores by two days of treatment.

Although there were no significant differences in sedation medications prior to screening positive for delirium, more patients in the antipsychotic-treated group had received benzodiazepines, opioids, and dexmedetomidine compared with patients in the untreated group

Adverse Events

No serious adverse events were reported in the treated group

Study Author Conclusions

InpharmD Researcher Critique

Inclusion and exclusion criteria were not mentioned in the study, making it hard to extrapolate this data. The patients who were treated with antipsychotics appeared to be sicker than the controls, so selection bias may have confounded these results.

Also, this is a retrospective study that relies on previous documentation for accurate retrieval of information. The primary endpoints of this study do not focus on which antipsychotic would be best for critically ill children experiencing delirium. 

Infant Delirium in Pediatric Critical Care Settings

Design

Case report

Case Presentation

A 7.5-month-old, 9.4 kg girl was hospitalized to treat stage IV neuroblastoma. She had extensive metastases in bone marrow, liver, lungs, and cranial bone involvement. The first cycle of chemotherapy was complicated due to constant agitation and respiratory compromise. The patient was sedated using fentanyl, hydromorphone, diphenhydramine, and lorazepam. 

Due to worsening of agitation, she was intubated and sedated using trials of sedatives such as propofol, phenobarbital, and pentobarbital, which did not improve her status. She maintained highly agitated when sedatives wore off. She was kept on ventilation for nine days in combination with sedatives. 

Psychiatry was consulted on the 11th day in the pediatric intensive care unit (PICU). The patient was extubated to receive the next round of chemotherapy. Upon examination, the patient was tachycardic, had facial edema, and was described by the mother as screaming and wiggling. The mother stated that the patient did not recognize her and was not relaxed by usual methods. 

The patient was assessed using the Pediatric Anesthesia Emergence Delirium Scale which was 19 (score of ≥10 indicated delirium). Electrocardiogram (ECG) showed no seizure activity. The patient was diagnosed with multifactorial delirium secondary to underlying disease. 

The patient was started on a trial of IV haloperidol 0.25 mg (0.025 mg x 10 kg) Q6H to achieve safe airway management, reduce stress, and reduce potential offending agents. The dose was determined by guidelines based on children >3 years of age (0.05-0.15 mg/kg divided into two or three doses per day). 

The patient's delirium and agitation improved within 24-hours. This allowed for phenobarbital to be slowly decreased and extubating of the patient occurred after the 3rd day of haloperidol initiation. The second cycle of chemotherapy was given and completed successfully. 

The patient developed "repetitive movements" on day five, which prompted evaluation for seizure, opiate withdrawal dyskinesia, and extrapyramidal symptoms. Psych recommended continuation of haloperidol at a reduced dose of 50% from initiation. The clinical team was concerned with possible extrapyramidal symptoms and decided to discontinue haloperidol. The psych team recommended the use of quetiapine 6 mg BID (1/4th of the recommended dose for children >13 years of age) for delirium due to a lower risk of extrapyramidal symptoms. The patient was cross-tapered over three days to quetiapine. 

The patient improved neurologically and medically following quetiapine initiation. The clinical team continued quetiapine throughout sedation tapers. Quetiapine was continued for a total of 6 weeks since initiation and slowly decreased to 6 mg QHS for two nights before being discontinued. The patient did not experience any adverse events. 

Study Author's Conclusions

Validation of the SOS-PD scale for assessment of pediatric delirium: a multicenter study

Design

Observational, multicenter, prospective, validation study

N=485

Objective

To validate the Sophia Observation Withdrawal Symptoms scale Pediatric Delirium scale (SOS-PD scale) in assessing pediatric delirium

Study Groups

Patients without confirmed delirium (n= 437)

Patients with confirmed delirium (n= 48)

Inclusion Criteria

Aged 3 months to 18 years old; admitted to a pediatric ICU; expected length of stay ≥48 hours

Exclusion Criteria

Anticipated death within 48 hours, neurological abnormalities, coma, or deep sedation

Methods

Patients that were assessed and met study criteria were evaluated via the PD scale (the pediatric delirium [PD] component of the SOS-PD scale) by research nurses TID (at times 04:00, 14:00, and 20:00) during their ICU stay. A score of ≥4 or had a positive item for hallucinations met the criteria for psychiatric consultation due to "suspected delirium".

The reliability of the PD-scale was assessed by comparing outcomes with similar outcomes of the Cornell Assessment Pediatric Delirium (CAP-D) tool. Validity was assessed by calculating the sensitivity, specificity, positive and negative predictive value, and the positive and negative likelihood ratios.

Duration

6 months

Outcome Measures

PD-scale sensitivity, specificity, reliability

Baseline Characteristics

Without confirmed delirium (n= 437)

Confirmed delirium (n= 48)

Age, years

3–24 months 

2–5 years 

5–12 years 

> 12 years

215 (49.2%)

70 (16.0%)

89 (20.4%)

63 (14.4%)

19 (39.6%)

8 (16.7%)

11 (22.9%)

10 (20.8%)

0.540

Female

Type of respiratory support 

None

Oxygen

Non-Invasive ventilation

Ventilation

HFO ventilation

61 (14.0%)

129 (29.5%)

15 (3.4%)

218 (49.9%)

14 (3.2%)

1 (2.1%)

2 (4.2%)

1 (2.1%)

40 (83.3%)

4 (8.3%)

Results

Validity 

Psychiatrist: delirium -

Psychiatrist: delirium +

No evaluation

PD scale ≥ 4

PD scale < 4

Total

41

1 

42

13 

113

126 

3 

314

317 

57

428

485

The sensitivity was 92.3% and the specificity was 96.5% after correction for verification bias, for a PD score ≥ 4.

The positive predictive and the negative predictive values were 76.4% and 99.1%, respectively.

Adverse Events

Study Author Conclusions

The SOS-PD scale has good inter-rater reliability and validity for screening by nurses of delirium in critically ill children

InpharmD Researcher Critique

The Use of Haloperidol in Agitated Critically Ill Children

Design

Case Report

Case Presentation

A 9-month-old female with metachromatic leukodystrophy developed a severe nosebleed. This led to intubation 14 days after a bone marrow transplant. After being on mechanical ventilation for a two-week period patient became difficult to adequately sedate;  10 µg/kg/hour of fentanyl, and 1 mg/kg/hour of midazolam was used (usual dosing1-2 µg/kg/hour, 0.05-0.15 mg/kg/hour) to induce sedation. The patient was subsequently reintubated but experienced subglottic narrowing.

After 33 days of mechanical ventilation another round of fentanyl was given at 30 µg/kg/hour, methadone 1.4 mg/kg IV every 6 hours, and lorazepam 1.1 mg/kg IV every 4 hours. To minimize muscular movement the patient was given 11 doses of pancuronium. Documentation noted the patient to be “agitated, thrashing”, and “struggling to achieve sedation”.

On day 33, at 10:30 pm 0.06 mg/kg of intravenous haloperidol was administered, 0.015 mg/kg IV was then given every 6 hours. The patient was noted to be calm and quiet 12.5 hours after the administration of the first dose and required two additional doses of haloperidol 0.015 mg/kg at 1:30 pm and 5:50 pm. Within the day following the loading dose with haloperidol, 1 dose of pancuronium and 2 additional doses of fentanyl were administered.

There were no more documented reports of agitation, and they were noted to be “easily calmed easily and "patient settles well with repositioning."  There were two additional notes that the "major problem was addressed for her by achieving sedation. Three days later, haloperidol was discontinued, other sedatives were titrated down, and the patient was successfully extubated.

Patient 1

Haloperidol dose

Load

Regimen

0.09 mg/kg

0.015 mg/kg q6h

0.2 mg/kg

0.15 mg/kg q8h

0.25 mg/kg

0.05 mg/kg 18h

0.2 mg/kg

0.1 mg/kg q8h

0.1 mg/kg

0.025 mg/kg q6h

Nursing notes

Before haloperidol

After haloperidol

Agitated

Calms easily

Agitated

Cooperative

Thrashing

Relaxed

Disoriented

Appropriate

Combatative

Well sedated

Adverse effects

Author Conclusions

In this case series, administration of intravenous haloperidol resulted in prompt behavioral control in 5 difficult-to-sedate critically ill children. While sedation scores were not used to assess patients in this series, optimal sedation in our unit is characterized as a lightly sleeping child with normal muscle tone.

This case series suggests that intravenous haloperidol may be a useful addition to the sedative armamentarium in carefully selected, agitated critically ill children. Its safety, efficacy, pharmacokinetics, and pharmacodynamics should be characterized in prospective studies.