Pediatric Delirium

Delirium, an acute transient disorder of global brain function, is becoming increasingly viewed as a psychiatric emergency in adult medicine. In adults, both the public health and individual patient burdens of delirium have proven to be extremely high. Delirium is present in 49% of all adult inpatient hospital days, resulting in an additional $16,303–$64,421 per delirious patient per year (1). Furthermore, the one-year risk of mortality associated with delirium in adults is 35%–40% (2). According to DSM-5, delirium is a disturbance in attention or awareness accompanied by changes in cognition that develops over a short period of time, fluctuates in course, and is the result of a medical condition (3). This definition of delirium can be difficult to apply in the pediatric setting, however, as a child's premorbid neurocognitive stage and language abilities must be taken into account. While neither DSM-5 nor ICD-10 includes a definition of delirium specific to pediatrics, multiple validated tools for assessing pediatric delirium have recently been developed. With these additional diagnostic tools, delirium is becoming increasingly recognized in children. Pediatric delirium comprises 10% of all pediatric consultation-liaison consults, occurs in up to 29% of critically ill children, and is a marker of serious illness, with an associated mortality rate of 20%. (4, 5). As noted by Schieveld and Janssen (6), timely recognition and treatment of pediatric delirium is necessary because the hypermetabolic state associated with delirium may impair recovery from critical illness, agitated behaviors associated with hyperactive delirium impede care, and the psychological effects may be traumatic.

Clinical Characteristics Of Pediatric Delirium

As in the adult population, delirium in the pediatric population can be classified based on the psychomotor state into hyperactive, hypoactive, or mixed delirium (7). While many of the clinical features of adult delirium can be applied to children, certain features are more prominent in children, which necessitate a unique approach to the pediatric delirium examination. For example, in a preverbal child the examiner might forgo formal bedside tests of attention and instead assess inattentiveness by observing poor eye contact or difficulty with engagement. Caregiver involvement can prove very helpful in making this diagnosis. Features of delirium that are particularly prominent in the pediatric population include irritability, affective lability, agitation, sleep-wake disturbance, and fluctuations of symptoms. In contrast, delusions, hallucinations, speech disturbances, and memory deficits are less commonly seen in children. Unique features of pediatric delirium include developmental regression with loss of previously acquired skills, inability of the usual caregiver to console the child, and reduced eye contact with the usual caregiver (8). Schieveld and Janssen (6) also described the “inconsolable child” as a red flag for delirium. A child who is agitated, breathing against the ventilator, and receiving escalating doses of sedating medications, should be considered to be delirious until proven otherwise (6).

Sequelae of delirium

Research on the sequelae of pediatric delirium remains in its infancy. Unlike in adults, research has not yet shown that an episode of pediatric delirium increases mortality independently of illness severity. However, Turkel et al. (4) demonstrated an increased length of hospital stay associated with a diagnosis of pediatric delirium. Furthermore, one-third of patients discharged from a pediatric intensive care unit meet criteria for posttraumatic stress disorder (PTSD) 3 months after discharge. Since PTSD symptoms can occur even in children who lack conscious memory of the trauma, delirium is a likely contributor to this finding (9). In the adult literature, it is becoming increasingly recognized that delirium can have negative long-term neurocognitive effects; however, it remains unclear whether this holds true for children. Early studies comparing generally medically hospitalized children to critically ill children show that critically ill children have impaired visual and spatial memory, as well as impaired attention, suggesting that an episode of pediatric delirium may affect long-term brain function (10).

Diagnosis

Pediatric delirium remains vastly underdiagnosed both by pediatric and psychiatric teams (11). There are a number of challenges associated with accurately and systematically diagnosing pediatric delirium. Because of inherent communication limitations in evaluating preverbal or nonverbal children, the diagnosis is contingent on close observation of behavioral symptoms. Additionally, the symptoms of pediatric delirium can be subtle, can vary depending on developmental stage, and are complicated by developmental variability. Involvement of the caregiver, who may not be easily accessible, is necessary to make a diagnosis. Finally, many of the symptoms used to make a diagnosis of pediatric delirium overlap with a number of other conditions, such as pain, distress, or drug withdrawal.

Fortunately, several delirium rating scales have recently been developed and validated. The Pediatric Anesthesia Emergence Delirium Scale, the Pediatric Confusion Assessment Method for the ICU, the Cornell Assessment of Pediatric Delirium, and the Sophia Observation Withdrawal Symptoms-Pediatric Delirium Scale comprise the four validated delirium screening tools for children. With comparable sensitivity (83%–94%), specificity (79%–98%), and feasibility of use, there is no clear “best” tool (12). However, the advantage in using a validated tool is increased rates of routine screening, thereby improving diagnostic accuracy and implementation of treatment.

The role of adjunctive tests to make the diagnosis of delirium remains limited. EEG shows diffuse slowing in only 65%–86% of pediatric cases, with fluctuations that parallel the clinical state (8). A number of candidate biomarkers, including hemoglobin-beta, S100 calcium-binding protein B, and IL-6 for delirium are being investigated; however, they are not routinely used to make the diagnosis. Thus, pediatric delirium fundamentally remains a clinical diagnosis.

Management

Inouye's (13) three-pronged approach for the management of adult delirium can readily be adapted for the pediatric patient. Her approach includes identifying and addressing predisposing factors and providing symptomatic care and symptom-targeted treatment. Firstly, all delirious patients should undergo a thorough assessment to identify the underlying cause of delirium, with special attention to the three most common causes of delirium in children: infection, medication-related factors, and autoimmune-related factors (4). The most common deliriogenic medications include anticholinergic agents, benzodiazepines, and opioids. These medications should be minimized, substituted, or tapered as medically appropriate. In the adult literature, there is increasing evidence for use of dexmedetomidine as an alternative, non-deliriogenic sedative agent; however, this has not been studied in children. Secondly, supportive care in delirium includes addressing volume and nutritional status, early mobilization, and deep venous thrombosis prophylaxis. Finally, delirium symptoms should be managed as they arise. Behavioral strategies can include frequent presence of the caregiver, having a familiar toy or photographs available, avoiding physical restraints, and normalizing the sleep-wake cycle.

Generally, pharmacologic intervention is recommended when the patient is distressed by the symptoms, the symptoms impose a safety concern, or they are impeding advancement of medical care (14). While there are no agents with Food and Drug Administration approval for delirium treatment in either adults or children, antipsychotics have been clinically shown to address delirium symptoms in adults and are widely used. Since there are few studies that examine the safety and efficacy of antipsychotics in pediatric delirium, the current psychopharmacologic approach to managing pediatric delirium is modeled on experience in adults.

Generally, while atypical antipsychotics are favored over typical antipsychotics in children, both are used, and no formal guidelines exist to guide antipsychotic selection. A retrospective study of 110 children aged 1–18 years diagnosed with delirium showed that delirium scores decreased significantly with atypical antipsychotic use (15). In the study, olanzapine and risperidone were used based on provider preference. Rates of delirium resolution were similar between the two antipsychotics. Few adverse side effects were observed: only one patient developed mild dystonia, which resolved quickly with dose reduction, and no cardiac or metabolic side effects were observed. Both quetiapine and intravenous haloperidol have also been shown to be generally safe and efficacious in managing pediatric delirium (16, 17). However, in a separate retrospective study of 26 acutely ill children who received haloperidol for hyperactive delirium, 23% experienced adverse reactions, including dystonia and hyperpyrexia (18). Another observational study of two cases of delirium in adolescent girls suggests that the particular subtype of delirium may predict a differential response to antipsychotics, with hyperactive delirium being more responsive to haloperidol and mixed/hypoactive delirium being more responsive to risperidone (7).

Preventive Strategies

In adults, delirium incidence can be robustly and safely reduced both by multi-component non-pharmacologic approaches and antipsychotic use (19, 20). In contrast, little has been published on delirium prevention strategies in children. It may be prudent to consider adapting adult preventative strategies for high-risk children.

Conclusions

Delirium, a syndrome of acute brain failure caused by medical illness, is becoming increasingly recognized in children. Although research in this field remains limited, early studies indicate that it is common, likely has negative long-term sequelae, and is treatable with both nonpharmacologic and pharmacologic approaches.

Key Points/Clinical Pearls