The Diagnosis and Management of Delirium in Infancy

Introduction

D elirium is a neuropsychiatric syndrome that indicates serious cerebral dysfunction (Rummans et al. 1995; Turkel 2003; Turkel et al. 2012). Etiology is often multifactorial, and more than one pathophysiologic mechanism may result in the characteristic symptoms of delirium (American Psychiatric Association 2000; Silver et al. 2010). Circulating cytokines and locally acting chemokines, impaired oxidative metabolism, abnormal neurotransmitter release, and alteration in complex neurotransmitter interactions may produce or exacerbate delirium in the susceptible patient (van der Mast 1998; Seaman et al. 2006; de Rooij et al. 2007).

Although the American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) criteria for the diagnosis of delirium are applicable in pediatric patients, delirium is usually recognized in adults and older children (American Psychiatric Association 1994; Turkel and Tavaré 2003; Creten et al. 2011). It is more difficult to make the diagnosis in the very young because of the need for experience in the evaluation of behavior and cognition in infants and toddlers (Turkel et al. 2012). Children with delirium may appear to have a more acute onset and more severe symptoms than adults (Leentjens et al. 2010), but this likely reflects the difficulty in accurately diagnosing younger patients (Turkel et al. 2012).

Regulation of state and attention appear to be fundamental central nervous system (CNS) functions, which are disrupted in delirium and are critical to the diagnosis, especially in infants and toddlers. The ability to mobilize, focus, and sustain attention can be easily appreciated in infants by noting how they engage, make eye contact, and remain interested. State regulation is a reflection of neurologic integrity, and assessment of state is based on eye opening and movements, gross body movements, muscle tone, and respiration. The four basic states are quiet awake, active or fussy awake, drowsy or active sleep, and quiet sleep (Holditch-Davis et al. 2004).

The characteristic core symptoms of delirium are recognizable in a patient of any age, and include impaired consciousness and awareness; the inability to direct, focus, sustain, and shift attention; abnormalities of the sleep–wake cycle; and disturbance of thought processes (Kean et al. 2010; Lishman 1987) The Delirium Rating Scale (DRS) assesses these basic symptoms of delirium. It does not rely only on higher order cognitive functions, which are difficult to assess in children <3 years of age, and it has been used to confirm the diagnosis in very young patients with delirium (Turkel 2003).

Instruments developed to confirm the diagnosis of delirium consist of operationalized criteria incorporating information from patient observation and the medical record. The DRS has been shown to have excellent reliability and validity (Trzepacz et al. 1988; Rockwood et al. 1996). The use of the DRS has been described in children and adolescents, with scores similar to those of adults (Turkel 2003).

The management of delirium is predicated on rapidly identifying and addressing its underlying cause, which usually results in the rapid alleviation of symptoms (Rummans et al. 1995; Grace and Holmes 2006). Supportive care with frequent reassurance and reorientation by family or a familiar nurse reduces fear and confusion, maintains a patient's connection to the environment, and may preclude the need for medication (Meagher et al. 1996; Inouye et al. 2003; Martini 2005).

When these important environmental interventions are insufficient, medication is frequently used to control agitation, diminish distress, and prevent harm (Meagher 2001; Martini 2005; Michaud et al. 2007). Antipsychotic medications, classically, parenteral haloperidol and more recently, atypical antipsychotics, effectively reduce the symptoms and hazards of delirium (Harrison et al. 2002; Boettger and Breitbart 2005; Stoddard et al. 2006; Alici-Evcimen and Breitbart 2008).

Medication administration to control symptoms and prevent potential adverse effects of delirium should be patient specific, goal oriented, and sensitive to potential risks in the critically ill pediatric patient, particularly the very young (Meagher 2001; Martini 2005). Although antipsychotics are considered to be the medications of choice, they are not approved by the Federal Drug Administration for management of delirium (American Psychiatric Association 2000; Maldonado 2008; DeMaso et al. 2009; Smith et al. 2009).

In order to expand appreciation of delirium in the very young, a retrospective study was undertaken to describe its presentation and the response to antipsychotic medication in infants and toddlers. The DRS was employed to confirm the diagnosis of delirium, characterize its presentation, and assess symptoms before and after administration of atypical antipsychotic medications.

Materials and Methods

This study was approved by the Children's Hospital Los Angeles (CHLA) Committee for Clinical Investigations (Institutional Review Board).

All psychiatric consultations for inpatients seen from January 1, 2004 through December 31, 2007 were reviewed, to identify patients with delirium who were <3 years old. The diagnosis of delirium was predicated on American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th ed., Text Revision (DSM-IV-TR) criteria and confirmed by the DRS (Trzepacz et al. 1988; American Psychiatric Association 2000). The DRS was also used to assess symptoms before and after administration of antipsychotic medication. No distinction was made for treatment of hypoactive, hyperactive, or mixed presentation of delirium. All patients were provided supportive and environmental interventions directed at restoring orientation, maintaining safety, and providing comfort.

Although the DRS was not scored at the time of evaluation, delirium symptoms were prospectively noted by the two attending child psychiatrists as part of the standard assessment, which facilitated later scoring. The DRS was retrospectively scored twice: When antipsychotic medication was initiated and when it was discontinued. Patient demographics, type of antipsychotic medication, dosage, symptoms, and pre- and post-antipsychotic DRS scores were compiled, and data were analyzed using a proprietary statistical package (Minitab 1984).

Results

Nearly half of all admissions to CHLA during the study period were children <4 years of age. Of the very young patients referred for psychiatric consultation who clinically met the diagnostic criteria for delirium during the study period, there were 10 males and 9 females, ages 7–34 months (mean age 20.5 months), and their diagnosis was confirmed by use of the DRS.

The most prominent symptoms found in these very young patients with delirium were impaired attention, sleep disturbance, irritability, and agitation (Table 1). Impaired state regulation and responsiveness were also noted, but not as often. All symptoms improved following administration of antipsychotic medication. The spectrum of symptoms associated with delirium in these very young patients was similar to that in older children (Turkel 2003; Turkel et al. 2012). Cognitive function and the sense of time and place are not well developed in infants and toddlers, and could not be used for diagnosis or rated in the DRS (Table 2). Perceptual disturbances, hallucinations, or delusions, were also difficult to appreciate in this mostly nonverbal population, although the parents of one patient described what appeared to be auditory hallucinations.

Olanzapine (n=16) and risperidone (n=3) were clinically chosen for the patients in the study. The starting dose of olanzapine was 0.625, or 1.25 mg q.h.s. or b.i.d. (Table 3), and that of risperidone was 0.05–0.1 mg q.h.s. or b.i.d. (Table 4). Relatively higher antipsychotic doses were required for patients in the intensive care units (ICU) who were receiving opioids and benzodiazepines. The dose was then titrated usually using control of insomnia and agitation as the target symptoms. Average duration of antipsychotic use, as reflected by the time between pre-and post-antipsychotic DRS score, was similar for both drugs (overall mean 38 days). The initial DRS scores (range 14–21, mean 16.8) were consistent with delirium (Trzepacz et al. 1988; Turkel 2003b). The DRS score at end of medication administration (range 4–10, mean 6.2) was significantly lower (p<0.0001) than the initial score (Table 5).

Patients were closely monitored while on antipsychotic medication. Antipsychotic dose was reduced as symptoms of delirium improved, and the antipsychotic was discontinued as soon as possible when sleep and behavior were normal. Three patients died of their underlying condition, representing a 16% mortality rate. Adverse side effects of antipsychotic use, including abnormal muscle tone, movement abnormalities, or cardiac arrhythmia, were not encountered with either antipsychotic agent.

Discussion

Antipsychotic medications are not approved by the United States Food and Drug Administration (FDA) to manage delirium in adults or children. Nonetheless, they are widely used because of their efficacy, and the lack of alternatives. When making a treatment recommendation, the theoretical risks of using psychotropic medication in very young patients must be considered against the risks for persistent cognitive impairment of untreated delirium (Prugh et al. 1980; Jackson et al. 2004). The need to address the symptoms of delirium, including significant agitation endangering lines or artificial ventilation, necessitates the use of methods developed for the diagnosis and management of delirium in older patients. Olanzapine and risperidone were clinically selected in this study because both are easy to administer orally and come in water- soluble oral disintegrating tablets, and risperidone is also available in liquid concentrate, allowing small doses to be administered.

Delirium has been reported in children and adolescents in association with CNS infections and sepsis, HIV/AIDS, cancer chemotherapy, closed head injuries, seizures, emergence from anesthesia, organ failure, hypoxia, fractures or trauma, or sedative-hypnotic and anticholinergic medications (Turkel 2003 and Tavaré; Schieveld et al. 2007; Turkel et al. 2012). The distribution of underlying medical diagnoses and mortality rate in the patients in this study is congruent with that in previous studies (Turkel and Tavaré 2003; Schieveld et al. 2007; Turkel et al. 2012).

Many of the patients with delirium were seen in the intensive care setting, where opioids and benzodiazepines were used to address agitation, especially in intubated patients (Tobias 2000; Stoddard et al. 2006). Refractory agitation in critically ill children in the pediatric ICU is likely a marker for pediatric delirium (Schieveld et al. 2010). Sedative and analgesic medications used in the ICU to relieve anxiety and pain associated with intubation may induce or exacerbate delirium (Breitbart 1996). Long-term sedation with opioids and benzodiazepines to control agitation and facilitate mechanical ventilation in the ICU often leads to development of tolerance and the need for dose escalation (Harrison et al. 2002). Escalating doses in turn contribute to confusion and agitation, resulting in delirium from both the underlying disorder and medications used to facilitate its treatment (Clegg and Young 2011). Antipsychotic use may facilitate lower sedative-hypnotic and narcotic use, providing an alternate option to address excessive agitation without exacerbating delirium.

All patients identified with symptoms of delirium during the study period received antipsychotic medication; therefore, there was not a comparable group of patients with delirium without antipsychotics to use as controls. It is not possible to conclude that the reduction in delirium symptoms, as reflected by lower DRS scores, was solely the result of the antipsychotic agent administered. Improvement in delirium symptoms would be anticipated as the underlying medical condition improved, and may reflect improved control of the factors predisposing to delirium, beneficial environmental interventions, or removal of medications exacerbating delirium (Tobias 2000; Stoddard et al. 2006). Nonetheless, antipsychotic medication use was associated with significant lowering of DRS score and improvement in symptoms associated with delirium.

Conclusion

Delirium, although rarely diagnosed in infants and toddlers, is similar in presentation to delirium in older patients. The symptoms of delirium in these very young patients responded to antipsychotic administration in a similar manner to those in older children, adolescents, and adults. Use of olanzapine and risperidone to control symptoms of delirium appears to be safe and effective in this population.

Clinical Significance

The management of delirium has been less studied in pediatric patients than in adults, and has been rarely addressed in the very young. This retrospective report describes the use of olanzapine and risperidone in managing the symptoms of delirium in infants and toddlers. Using the DRS to confirm the diagnosis and assess response, both agents were found to be useful (p<0.0001) without significant adverse side effects. This study provides data to inform further research, and supports the clinical supposition that atypical antipsychotics can be helpful in the management of very young pediatric patients with delirium.