This article contributes to a series of updates for the fifth edition of the Rogers’ Textbook of Pediatric Intensive Care. The overall objective of this targeted review is to update readers about key recent research that has furthered our understanding and management of pain, sedation, sleep, and delirium in the PICU since the publication of the fifth edition. The last 5 years have brought exciting new information about pain and sedation assessment and protocols, safety and efficacy of sedative-analgesic regimens, reversal of neuromuscular blockade (NMB), and sleep in the PICU. Additionally, there has been extraordinary progress since the fifth edition to elucidate the epidemiology and risk factors for delirium in critically ill children in a wide spectrum of disease processes.
Given decreasing PICU mortality rates and increasing morbidity rates, the focus of PICU care is shifting toward reducing iatrogenic harm as a result of PICU care and improving long-term outcomes for patients (1). Thus, an area of intense focus is long-term sedative and analgesic exposure in mechanically ventilated patients and the risks associated with benzodiazepine use in particular as it relates to delirium. When the fifth edition was published, validated screening tools had just been established for recognition of delirium in critically ill pediatric patients, and this review will provide updates on what we have learned thanks to the availability of these tools. Since publication of the fifth edition, our field has also learned from the largest clinical trial of sedation protocolization in critically ill children.
The ICU Liberation movement and “ABCDEF” bundle (2), a multidisciplinary and integrated approach to improve outcomes in critically ill patients, has established the interconnection between assessing, treating, and managing pain (A), choice of sedation (C), and delirium monitoring and management (D) (3). Reviewing new literature on both spontaneous awakening and extubation readiness trials (B), early mobilization (E), and family engagement (F) is outside the scope of this update, but it is important to consider how we may approach pain, sedation, and delirium management within the context of ICU Liberation. At the time of this update, the most recent clinical practice guidelines from the Society of Critical Care Medicine have added immobility and sleep disruption to pain, agitation/sedation, and delirium as critical considerations for clinical care and research in the adult ICU (4). Although ICU Liberation began in adult ICUs, these same ABCDEF principles are being progressively addressed in pediatric critical care. While we await pediatric-specific guidelines, the literature included in this review highlights increased attention by our field to each of these areas and how their interplay impacts short- and long-term outcomes in infants and children.
Benzodiazepines have been established as an independent risk factor for delirium development in critically ill infants and children. Thus, data regarding the use of medications other than opiates and benzodiazepines for sedation are evolving and comprise a substantial proportion of the recent literature. Medications such as alpha-2 agonists and ketamine are gaining popularity for use during procedural sedation, as well as for longer term sedation management of the PICU patient. Current evidence demonstrates their safety and efficacy when used at appropriate dosages. A major advance in NMB reversal has emerged with sugammadex. Emerging research in the PICU is characterizing sleep in critically ill children and staff perceptions of the PICU environment for sleep hygiene. Finally, we are gaining new insights about the prevalence and cost of delirium and its impact in different subsets of PICU patients, including those undergoing cardiac surgery and those receiving extracorporeal membrane oxygenation (ECMO).
PAIN AND SEDATION ASSESSMENT
Critically ill children commonly experience pain secondary to trauma, medical procedures, invasive devices, and illness-induced discomfort. Assessment of pain in pediatrics is challenging because of variability in cognition and development, difficulty with communication, and the complex interplay between sedative and analgesic medications. Pain is a subjective measure, and efforts to standardize assessments using objective, noninvasive monitoring such as bispectral index or cutaneous conductance have been unsuccessful (5 , 6).
Bedside nurses play a central role in assessing the pain of children, and current recommendations encourage the use of validated pain scales to guide alleviation of pain (7). The COMFORT Scale; Face, Legs, Activity, Cry and Consolability; and the Multidimensional Assessment of Pain Scale are the three most commonly used pain assessment tools in the PICU setting and are recommended by the European Society of Paediatric and Neonatal Intensive Care (8). Recent studies, however, suggest that bedside nurses do not consistently use these assessment tools. In a study that used virtual humans and written vignettes, PICU nurses used behavior as a primary indicator to assess and treat pain, even when the child was mature enough to articulate (9). PICU nurses exhibit marked variability in their assessment, beliefs, and response to pain in critically ill children. Using both quantitative and qualitative tools, Lafond et al (10) found that many undertreat pain out of concern for adverse effects of opioids or knowledge gaps regarding pharmacokinetics.
These recent studies highlight the wide variation in nursing assessment of pain and emphasize the need for standardization. Keogh et al (11) showed that using practice guidelines for analgesic management in critically ill children was feasible and led to an improvement in staff’s understanding of pediatric pain and its treatment. Implementation of the guidelines did lead to higher minimum and maximum dosages of morphine and midazolam, but lower average length of continuous infusions, suggesting that the pain was adequately controlled with minimal risk of withdrawal. Future directions in pain assessment of critically ill children should continue to focus on improving consistency in assessment and alleviation of pain.
Sedation should be goal-directed and customized for each patient with consistent and reliable assessment methods. Since the publication of the fifth edition of the Rogers’ Textbook of Pediatric Critical Care, two additional sedation scales have been validated for use in children: the Richmond Agitation Sedation Scale (RASS) and the Pediatric Sedation State Scale (PSSS).
The RASS is a scoring system commonly used in adults that incorporates both agitation and sedation. It had not been previously validated for the pediatric population. The RASS is unique because of its ability to assess awareness, which is important in the recognition of hypoactive and hyperactive delirium. Kerson et al (12) tested this tool in 100 patient encounters and from 50 unique patients aged 2 months to 21 years. Twenty-seven percent of the assessments were performed in mechanically ventilated patients; the remainder of the children were breathing spontaneously. This criterion for inclusion is distinctive, as the commonly used State Behavioral Scale is validated only in mechanically ventilated children (4). The RASS score was compared with both a Visual Analog Scale (VAS), given by the bedside nurse for assessment of agitation, and the University of Michigan Sedation Scale (UMSS), which was conducted by a researcher. The researchers found that the RASS was highly correlated with both the VAS (Spearman correlation coefficient, 0.8; p < 0.0001) and with the UMSS (weighted kappa, 0.9; p < 0.0001).
The PSSS is a six-point scale developed to measure the effectiveness and quality of procedural sedation, including the control of pain, anxiety, movement, and adverse side effects in pediatric patients. The PSSS was shown to have respectable interrater and intrarater reliability as well as validity when compared with the Observational Scale of Behavioral Distress-Revised scale (13). The scoring system differs from other sedation scales because it allows for analysis of procedural conditions regardless of presence or absence of sedation. For example, this scale can be used when performing a lumbar puncture in a pediatric patient with distraction instead of pharmacologic sedation, and will alert the caregiver of any dangerous conditions such as unsafe movements.
The development of the PSSS and the validation of the RASS scoring system will help PICU providers to administer optimal sedation and evaluate sedation status of patients who receive short-term procedural sedation or longer term sedation with or without an endotracheal tube.
SEDATION: THE ROLE OF PROTOCOLS AND SEDATION INTERRUPTION
Sedation is a central component of PICU care meant to ensure the comfort and safety of our patients. However, prolonged use of sedation can lead to detrimental side effects such as respiratory depression, constipation, tolerance, and physical dependence that can cause iatrogenic withdrawal when sedation is no longer needed. Reports in the adult ICU literature are conflicting with regard to improvement in mechanical ventilation time and length of stay with the use of protocolized sedation (5). Studies evaluating this topic in PICUs were not available at the time of the publication of the fifth edition of the Rogers’ Textbook of Pediatric Critical Care (14).
The Randomized Evaluation of Sedation Titration for Respiratory Failure (RESTORE) trial was an unblinded and clustered randomized control trial involving 2,449 mechanically ventilated children in 31 U.S. PICUs from 2009 to 2013 (15). PICUs randomized to the study group used a sedation protocol that incorporated arousal assessments, extubation readiness tests, and adjustment and weaning of sedation as needed per assessment. Control PICUs managed sedation per usual care without any protocolization. The RESTORE study showed no difference between the two groups in the duration of mechanical ventilation, inadequate pain or sedation management, iatrogenic withdrawal, or unplanned endotracheal or invasive catheter removal. In exploratory analysis, the authors did find that patients in the intervention group had fewer days of opioid administration (median 9 vs 10 d) and were exposed to fewer classes of sedative medication (median 2 vs 3) than the control groups. The intervention group was awake and calm 86% of the time compared with 75% of the time for the control group (p = 0.004); however the intervention group had more days with a pain score greater than 4 (60% vs 40%; p ≤ 0.001). Investigators also found a higher occurrence rate of postextubation stridor (7% vs 4%; p = 0.03) but no difference in reintubation rates, and a lower risk of pressure ulcers (< 1% vs 2%) in the intervention group. The authors concluded that the use of a nurse-implemented, goal-directed sedation protocol did not reduce the duration of mechanical ventilation and acknowledged that the relationship between wakefulness, pain, and agitation is complex.
A subsequent analysis from the RESTORE study that evaluated postdischarge outcomes showed no difference in postdischarge morbidity between patients who received a goal-oriented sedation protocol and those who received usual care (16). This study highlighted a significant physical morbidity in patients recovering from acute respiratory failure, with 19% experiencing a persistent decline in functional status 6 months after discharge. Although there was no observed physical benefit in the sedation protocol arm, the authors point out that there was also no difference in emotional dysfunction. Therefore, they concluded that a sedation strategy allowing patients to be more awake was not harmful from a physical or emotional perspective.
Daily sedation interruption was separately evaluated in a multicenter randomized control trial in the Netherlands by Vet et al (17). This trial was conducted from 2009 to 2014 and included mechanically ventilated children in three PICUs. These patients, who were randomized to receive protocolized sedation with daily sedation interruptions or protocolized sedation only, exhibited no difference in length of mechanical ventilation, cumulative midazolam dose, or hospital length of stay. The group without the daily sedation interruption did require more reintubations (3% vs 14%; p = 0.03). Interestingly, the authors did find a higher occurrence rate of mortality in the daily sedation interruption group (9% vs 0%; p = 0.03). The findings of this study differ from two earlier studies on daily sedation interruption in PICU patients (18 , 19). These previous studies showed that sedation interruptions led to diminished use of sedative medications, shorter mechanical ventilation time, and shorter ICU stays. In the most recent study by Vet et al (17), all patients were on a sedation protocol, whereas the previous studies used no sedation protocol. The authors speculate that the use of protocolized sedation itself outweighs the previously seen benefits of a daily sedation interruption.
Although these two new additions to the literature did not provide us clear supporting evidence for the benefit of sedation protocolization, they do highlight the shift in thought paradigm regarding sedation management in the PICU. The new target has evolved to frequent tailoring of medications to allow patients to be as awake and communicative as possible while balancing the need for pain management and anxiolysis. Each individual PICU must determine the best approach to achieve this goal in the setting of their own unit culture, whether through nurse-driven sedation protocols or frequent bedside assessments by the care team. All in all, each patient’s sedation plan should be individualized and titrated for his/her own clinical needs.
SEDATIVE MEDICATIONS IN THE PICU
Sedation of the critically ill child can be a daunting task, especially when attempting to achieve a balance among adequate sedation, anxiolysis, and pain control while minimizing short- and long-term effects from these centrally acting medications. Substantial research has been conducted to target the utilization of medications other than benzodiazepines and opiates for PICU sedation since publication of the fifth edition. We highlight the most impactful studies in the following sections.
Ketamine, a N-methyl-D-aspartate receptor antagonist, is a dissociative anesthetic that also provides analgesia with minimal respiratory depression and stable hemodynamics. These properties have increased its popularity for use in procedural sedation. Previous studies of ketamine’s safety have been limited to the pediatric emergency department. The Pediatric Sedation Research Consortium (PSRC) published data on 22,645 sedations with ketamine outside of the operating room from 2007 to 2015 (20). The locations of the sedation events were primarily in sedation or radiology suites (65%), with 12% occurring in the emergency department and 7% in the PICU. Ketamine was used as the sole medication in 17% of the cases; benzodiazepines and propofol were the most commonly coadministered agents (58% and 35%, respectively). The authors found a low occurrence rate of serious adverse events (SAEs) (2%) and an overall adverse event (AE) rate of 7%. No patients died, although three in radiology/sedation suites had cardiac arrests, all related to laryngospasm. Dental procedures comprised a small proportion of the cases (0.6%) but did have a significantly higher risk of SAE. The coadministration of benzodiazepines was associated with decreased odds of SAE and AE. Coadministration of propofol, anticholinergics, or barbiturates was associated with significantly higher percentages of AEs and SAEs. The authors also found a dose-dependent relationship. Dosages greater than 2.5 mg/kg were associated with a significant increase in AE (7% vs 5%; p = 0.008), and dosages greater than 5 mg/kg were associated with a statistically higher percentage of AEs and SAEs. The dosage-based risk factor is in agreement with a meta-analysis on ketamine administered in the pediatric emergency department that found a total dose of greater than 5 mg/kg to be associated with adverse respiratory events (21).
The PSRC later performed a separate analysis of the outcomes associated with the use of ketamine and propofol in pediatric patients outside of the operating room (22). The cohort included 7,313 procedural sedations that occurred mainly in dedicated sedation or radiology units. Anticholinergics and benzodiazepines were coadministered 14% and 41% of the time, respectively. AEs occurred in 10% of sedations, and the SAE rate was 3%. This SAE, although low, is higher than that for use of propofol (2%) or ketamine alone (2%). There were no deaths and one cardiac arrest, and airway obstruction was the leading cause of SAE. The authors also found higher odds of AE with the coadministration of anticholinergics, ASA status greater than or equal to III, and the primary diagnosis of a gastrointestinal illness.
Studies of ketamine use exclusively in the PICU are lacking. However, the safety profile of ketamine use in other high-risk units (emergency department, sedation suite) depicts a low rate of SAEs, with laryngospasm being the most common event. Therefore, ketamine use in the PICU setting has a substantial role in procedural sedation and even sedation for mechanical ventilation specifically in cases of severe bronchospasm or as an adjunct for children who are difficult to sedate. However, the interaction of ketamine with other agents, including propofol, must be considered with caution.
Alpha-2 Adrenergic Agonists
The alpha-2 agonists clonidine and dexmedetomidine have become increasingly used for sedation in the PICU over the last decade (23). This class of medications mediates sleep, analgesia, and sedation but does not provide amnesia. Alpha-2 agonists are an attractive choice for sedation because they do not decrease respiratory drive. The disadvantage of this class is the possibility for bradycardia and hypotension, which may limit these medications as a first-line choice for sedation, particularly in young infants and cardiac patients. Studies published since the fifth edition have focused on the safety profile of these drugs and indications for more widespread use.
Clonidine is available in multiple preparations, including nasal spray, oral liquids, oral tablets, oral transmucosal system, and rectal suspensions and is approved for IV use in some countries. A systematic review in which Hanning et al (24) compared the bioavailability and efficacy among the different formulations found the IV form to have the most predictable bioavailability. Given that clonidine has a long half-life of 12–24 hours owing to its large volume of distribution, down titration can be difficult with any formulation. The authors concluded that additional work needs to be done on oral, nasal, and rectal suspensions.
Studies on the efficacy of clonidine as an adjunctive sedation agent are limited. Hayden et al (25) performed a systematic review to evaluate the efficacy of alpha-2 agonists for sedation in the PICU. They found six randomized control trials that compared the use of an alpha-2 agonist with either another medication or a placebo. Three of the trials involved clonidine—two which compared it with placebo and one which compared it with midazolam. The authors found that clonidine had an opiate- and benzodiazepine-sparing potential when used as an adjunctive sedation in the neonatal age group only. This potential benefit was not supported in older children and was not associated with a decrease in the duration of mechanical ventilation in any age group.
Given clonidine’s long half-life, the potential for alterations in hemodynamics has been a concern, particularly in neonates. Kleiber et al (26) compared the use of an IV infusion of clonidine to midazolam as an adjunct to morphine in neonates after cardiac surgery. Although the patients who received clonidine had a transient decrease in diastolic blood pressure 13% of the time and a maximal decrease in heart rate of 12%, the neonates showed no signs of decreased cardiac output. Thus, these changes seemed of minimal clinical importance. The authors concluded that the use of IV clonidine appears hemodynamically safe in the neonatal population, although a larger study is warranted. Most recently, Kleiber et al (26) followed their neonatal study with an evaluation of hemodynamic tolerance of clonidine in a broad population of PICU patients ages 0–18 years (27). They concluded that although clonidine administration was often associated with bradycardia and hypotension, those complications were rarely clinically significant despite a high degree of illness severity.
Dexmedetomidine, an alpha-2 agonist that was approved in 1999 for use in adults, has gained popularity in pediatrics given its continuous infusion form, titratability, and short half-life. At this time, dexmedetomidine is not approved for use in children in any country; however its off-label usage is increasing, particularly in the United States. The PSRC performed a retrospective review to evaluate the safety of dexmedetomidine use for procedural sedation (28). Five percent of all sedations during the study period used dexmedetomidine. Of those, dexmedetomidine was coadministered with benzodiazepines in 61% of cases, ketamine in 6% of cases, and opioids in 4% of cases. Most uses of dexmedetomidine for sedation were for radiologic imaging, given its limited analgesic properties. The overall AE rate was 4%, and the SAE rate was a very low 0.34%, with airway obstruction being the most common. The rate of unexpected change in heart rate or blood pressure of greater than 39% was 0.93%, and there were no deaths or cardiac arrests in the patients who received dexmedetomidine. The occurrence rate of clinically significant hypotension or bradycardia is reassuring and supported by other studies. For example, a meta-analysis that evaluated the occurrence rate of bradycardia with dexmedetomidine use showed a wide range of 0–22% across previous studies, with an average of 3% occurrence rate after meta-regression analysis (29).
Dexmedetomidine can be suitable in some settings as a primary sedative agent and may be particularly useful as a sole agent when short- or long-term sedation is required in nonintubated patients. A recent review evaluating the use of intranasal dexmedetomidine for preprocedural sedation showed that, compared with oral benzodiazepines, dexmedetomidine had a longer onset of effect but led to better overall sedative effects without respiratory depression (30). Data from the RESTORE trial supports its use in low criticality patients (31), and Venkatraman et al (32) demonstrated dexmedetomidine to be effective as a single agent for sedation to facilitate tolerance of pediatric noninvasive ventilation. During noninvasive ventilation, the sole use of dexmedetomidine enabled the targeted sedation level in 83% of patients. Clinically significant hemodynamic changes were minimal and were responsive to a decrease in dexmedetomidine infusion, fluid bolus, or titration of the noninvasive ventilation. Additional studies that evaluated prolonged use of dexmedetomidine as a primary agent in noninvasive ventilation revealed predictable hemodynamic effects, with patients experiencing bradycardia and systolic hypertension during the escalation phase (33). This retrospective review by Shutes et al (33) also highlighted the iatrogenic withdrawal potential, as 25% of patients who received dexmedetomidine for more than 96 hours experienced withdrawal. The authors recommended the initiation of clonidine in this population.
A systematic review showed that when dexmedetomidine is used as an adjunctive medication, it can lead to decreased opioid administration; however this conclusion has not been supported in all trials (25). The RESTORE trial concluded that dexmedetomidine does not appear to have any added benefit as a secondary agent and offers inadequate pain control and sedation in these patients (31).
Given the absence of respiratory depression with dexmedetomidine use, clinical benefits from improved respiratory strength have been suggested. A Cochrane review showed that dexmedetomidine reduced the duration of mechanical ventilation and ICU stays in adult patients, and recent pediatric literature supports this notion as well (34). A study evaluating the use of dexmedetomidine in children with acute respiratory failure showed that it might prove beneficial as a periextubation agent by shortening the ventilator weaning process (31). In the pediatric congenital heart disease population, a meta-analysis of outcomes showed that dexmedetomidine use was associated with a shorter duration of mechanical ventilation, reduced stress response, and lower risk of delirium (35).
Reversal of NMB is often required to prevent postoperative residual paresis and can facilitate quicker extubation times. Previously, acetylcholinesterase inhibitors were the only reversal agents available and were associated with muscarinic side effects and residual blockade in both adults and children (36).
Sugammadex, a modified gamma cyclodextrin, was approved by the Food and Drug Administration in December of 2015 for the reversal of steroidal nondepolarizing NMB agents such as rocuronium and vecuronium (37). This is a clinically important development, as sugammadex acts more rapidly and effectively than neostigmine for full NMB reversal and does not have the associated muscarinic side effects that come with neostigmine. Sugammadex exerts its effect by forming a tight water-soluble complex with the neuromuscular agent. It encapsulates the drug particles and prevents their binding to nicotinic receptors at the neuromuscular junction (38 , 39). A meta-analysis of sugammadex use in pediatric patients was performed in 2016 (40). This review evaluated six randomized control trials that included a total of 253 pediatric patients and compared reversal by sugammadex, neostigmine, or placebo. The authors found that when compared with placebo or neostigmine, sugammadex shortened the rocuronium-induced NMB time, leading to faster extubations. There was no difference between the two groups in the number or severity of AEs. The dosage of sugammadex is dependent on the patient’s train of four at the time of administration. Limited data are available for children under 2 years old. The dosage chart is included in Table 1.
Because sugammadex is cleared renally, its use is not recommended for a patient with creatinine clearance less than 30 mL/min. If readministration of rocuronium or vecuronium is required after sugammadex administration, a clinical effect may not be observed for 5 minutes to 24 hours; therefore, it is advised instead to use a nonsteroidal neuromuscular blocking agent. The development of sugammadex is momentous for select PICU patients who require rapid and effective reversal, such as in the postoperative setting to facilitate extubation or for rescue in the “can’t intubate, can’t ventilate” patient.
WEANING OF OPIOIDS AND SEDATIVES
Tolerance, dependence, and then subsequent withdrawal from sedative medications are an iatrogenic complications of many critically ill children. Despite this common clinical problem, there has been little consensus on the best practice approach for weaning of sedative medications in long-term mechanically ventilated children. Sanchez-Pinto et al (41) evaluated the effects of an opioid weaning protocol on the opioid drug burden in PICU patients. Among 107 children (68 pre intervention and 39 post intervention), they found that the use of an opioid weaning protocol led to fewer days on opioids (23 d vs 17; p = 0.01) and a decrease in the total cumulative opioid exposure without an increase in withdrawal symptoms. These findings have been supported by previous studies (42), and the implementation of weaning protocols for opiates should be considered in all PICUs.
Of interest, the study by Sanchez-Pinto et al (41) did incorporate the novel approach of establishing a baseline Withdrawal Assessment Tool (WAT)-1 score as a key component of the weaning protocol. The baseline WAT-1 score allows for bedside caregivers to differentiate between withdrawal and overlapping clinical symptoms that could mimic and be misinterpreted as withdrawal thus leading to a prolonged wean. Examples of these symptoms include baseline hypertonia, hyperthermia, and persistent emesis. Recognizing and incorporating an individual patient’s baseline symptoms can likely facilitate a shorter duration of sedative weaning.
SLEEP IN THE PICU
Restorative sleep is important for neuronal development, metabolism, and immune system function; however previous studies have demonstrated significant sleep disturbances in the PICU (43–45). A prospective study recently evaluated temporal characteristics of the sleep electroencephalogram (EEG) in mechanically ventilated PICU patients and confirmed a lack of the normal ultradian variation and decreased slow wave sleep in critically ill children compared with that in healthy children (46). In that study, eight PICU patients with respiratory failure who required mechanical ventilation underwent limited montage EEG. The EEGs of the PICU patients were compared with those of eight age- and gender-matched healthy children. The typical patterns in δ or θ power spectral bands observed in the healthy subjects were entirely lacking in the PICU patients. Importantly, these children exhibited a behavioral state similar to sleep (eyes closed, resting) while having an EEG that did not support restorative sleep.
The lack of organized and restorative sleep in the PICU is multifactorial. Contributors include many modifiable factors, such as medication choice, nighttime interventions, and environmental stimuli. Single patient rooms allow for a shelter against noise and light outside of the room as well as privacy from a neighbor’s interventions. From a nurse’s perspective, the use of private rooms is more conducive to promoting a good sleep environment for patients than the use of multipatient rooms (47). Future research is needed to evaluate the interplays among the ICU environment, medication choices, patient activity, and sleep in pediatric critically ill patients.
DELIRIUM IN THE PICU
Delirium, characterized by a fluctuating disturbance in awareness, attention, and cognition, has been established as a significant morbidity in the PICU (35–38). A recent multinational point prevalence study found the prevalence of delirium in critically ill children to be notably high at 25% (48). Prolonged ICU stay of more than 6 days and mechanical ventilation requirement increased the prevalence of delirium significantly to 38% and 53%, respectively. Patients in cardiac intensive care exhibit an even higher prevalence, with one study showing that the occurrence rate of delirium was 100% in cardiac ECMO patients and 49% in children who had undergone surgery requiring cardiac bypass (49 , 50). Reported outcomes of patients with delirium are concerning. Recent studies have found that the diagnosis of delirium is associated with a prolonged ICU stay and is a strong and independent predictor of mortality (51). Pediatric delirium can put a substantial burden on healthcare costs. Delirium has been associated with an 85% increase in PICU costs, with the cost increasing incrementally with the number of days spent delirious (52).
Recognizing delirium in the pediatric population can be challenging given the wide diversity of ages, developmental stages, and mix of hypoactive and hyperactive delirium. At the time of publication of the fifth edition, the only validated tool for delirium screening in children less than 5 years old was the Cornell-Assessment of Pediatric Delirium (CAPD). The pediatric Confusion Assessment Method for ICU was previously validated in critically ill children greater than 5 years old and has been shown to have superior validity when compared with the Paediatric Anesthesia Emergence Delirium Scale (53 , 54). The Preschool Confusion Assessment Method for the ICU (psCAM-ICU) was most recently validated for use in children 6 months to 5 years old (55). The psCAM-ICU screening tool encompasses colored and mirrored cards to evaluate attention and can detect both hypoactive and hyperactive delirium. Given that most children admitted to the PICU are under the age of 5 years, having two validated tools, the psCAM-ICU and CAPD, is crucial for accurate and efficient delirium screening. Implementation of an ICU bundle that includes delirium screening, detection, and treatment has been shown to be feasible and effective, decreasing delirium rates from 17% to 12% in one PICU (56).
In light of the emerging literature regarding the high occurrence rate of delirium in our patients and the concern for associated worse clinical outcomes, we must be cognizant of and address modifiable risk factors. The delirium point prevalence study by Traube et al (48) showed an increased risk of delirium with younger age (< 2 yr), need for vasoactive infusions, and the need for antiepileptic medication, all factors that are nonmodifiable. However, the study also pointed out many iatrogenic risk factors. The use of benzodiazepines, narcotics, and physical restraints was all strongly associated with delirium. The concern for an association between benzodiazepine use and subsequent delirium has heightened.
A recent retrospective study by Mody et al (57) found benzodiazepines to be an independent predictor of delirium in critically ill children. They reported a temporal, causal, and dose-dependent relationship between benzodiazepine exposure and the development of delirium. Receipt of benzodiazepines more than doubled a child’s risk factor for being delirious the following day (odds ratio [OR], 2.0; p < 0.001), even after controlling for prior delirious status. In children who were not yet delirious, benzodiazepine exposure quadrupled the chance of delirium (OR, 4.4; p ≤ 0.002), showing these drugs to be independently associated with a transition from normal mental status to delirium. Similarly, analysis of data from the psCAM-ICU validation study of preschool age children (58) showed that benzodiazepine exposure was independently associated with the development of delirium (OR, 2.47; p = 0.005). In children greater than 12 months old, both benzodiazepine exposure and delirium were significantly associated with a lower likelihood of ICU discharge (hazard ratio, 0.65; p = 0.011). The conclusions from these recent studies suggest that a benzodiazepine-sparing approach may decrease the burden of delirium in our PICU population.
At the time of this review, the most recent literature had explored risk factors for delirium in postoperative patients admitted to the PICU (59). Notably, total IV anesthesia was associated with a lower risk of delirium development (p < 0.05). Additionally, a nested retrospective cohort study demonstrated an independent association between blood transfusion and delirium development (adjusted OR, 2.16; 95% CI, 1.38–3.37).
Another important and logical focus of the recent delirium literature is the approach to treatment once delirium has been diagnosed. Although the importance of using nonpharmacologic therapies as a first-line tactic is emphasized, the role of child psychiatry engagement and the safety and efficacy of antipsychotic therapy are growing areas of interest. In a case series describing child psychiatry consultation in cases of delirium, Barnes et al (60) found that over half of patients were started on an antipsychotic medication before involvement of child psychiatry. However, the involvement of the psychiatry team facilitated adjustment and weaning of medications after discharge from the PICU. Slooff et al (61) reported first results from prospective, systematic monitoring of AEs in 13 patients with confirmed delirium who were treated with a haloperidol dose titration protocol. Their study revealed a significant proportion of potential AEs despite low plasma concentrations and dosing within recommended ranges.
Significant advances have been made over the past few years in our understanding of the epidemiology, recognition, associated risk factors, treatment, and troublesome outcomes for patients with delirium. Interventions to mitigate harm to our patients by minimizing modifiable risk factors for delirium should be advocated. In particular, an ICU delirium bundle should be implemented universally, and the use of benzodiazepines and restraints, although many times medically necessary, should be limited when possible.
We would like to thank Claire Levine, MS, in the Department of Anesthesiology and Critical Care Medicine at Johns Hopkins for providing editorial assistance for this article.
1. Pollack MM, Holubkov R, Funai T, et alEunice Kennedy Shriver National Institute of Child Health and Human Development Collaborative Pediatric Critical Care Research Network: Pediatric intensive care outcomes: Development of new morbidities during pediatric critical care. Pediatr Crit Care Med 2014; 15:821–827
2. Marra A, Ely EW, Pandharipande PP, et alThe ABCDEF bundle in critical care. Crit Care Clin 2017; 33:225–243
3. Ely EWThe ABCDEF bundle: Science and philosophy of how ICU liberation serves patients and families. Crit Care Med 2017; 45:321–330
4. Devlin JW, Skrobik Y, Gélinas C, et alClinical practice guidelines for the prevention and management of pain
, immobility, and sleep
disruption in adult patients in the ICU. Crit Care Med 2018; 46:e825–e873
5. Solana MJ, Lopez-Herce J, Fernandez S, et alAssessment of pain
in critically ill children
. Is cutaneous conductance a reliable tool?. J Crit Care 2015; 30:481–485
6. Coleman RM, Tousignant-Laflamme Y, Ouellet P, et alThe use of the bispectral index in the detection of pain
in mechanically ventilated adults in the intensive care unit: A review of the literature. Pain
Res Manag 2015; 20:e33–e37
7. Harris J, Ramelet AS, van Dijk M, et alClinical recommendations for pain
, withdrawal and delirium
assessment in critically ill infants and children
: An ESPNIC position statement for healthcare professionals. Intensive Care Med 2016; 42:972–986
8. Harris J, Ramelet A-S, van Dijk M, et alClinical recommendations for pain
, withdrawal and delirium
assessment in critically ill infants and children
: An ESPNIC position statement for healthcare professionals. Intensive Care Med 2016; 42:972–986
9. LaFond CM, Vincent CVH, Corte C, et alPICU nurses’ pain
assessments and intervention choices for virtual human and written vignettes. J Pediatr Nurs 2015; 30:580–590
10. LaFond CM, Van Hulle Vincent C, Oosterhouse K, et alNurses’ beliefs regarding pain
in critically ill children
: A mixed-methods study. J Pediatr Nurs 2016; 31:691–700
11. Keogh SJ, Long DA, Horn DVPractice guidelines for sedation
and analgesia management of critically ill children
: A pilot study evaluating guideline impact and feasibility in the PICU. BMJ Open 2015; 5:e006428
12. Kerson AG, DeMaria R, Mauer E, et alValidity of the Richmond Agitation-Sedation
Scale (RASS) in critically ill children
. J Intensive Care 2016; 4:65
13. Cravero JP, Askins N, Sriswasdi P, et alValidation of the Pediatric Sedation
State Scale. Pediatrics 2017; 139:e20162897
14. Brook AD, Ahrens TS, Schaiff R, et alEffect of a nursing-implemented sedation
protocol on the duration of mechanical ventilation. Crit Care Med 1999; 27:2609–2615
15. Curley MA, Wypij D, Watson RS, et alRESTORE Study Investigators and the Pediatric Acute Lung Injury and Sepsis Investigators Network: Protocolized sedation
vs usual care in pediatric patients mechanically ventilated for acute respiratory failure: A randomized clinical trial. JAMA 2015; 313:379–389
16. Watson RS, Asaro LA, Hertzog JH, et alRESTORE Study Investigators and the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network: Long-term outcomes after protocolized sedation
versus usual care in ventilated pediatric patients. Am J Respir Crit Care Med 2018; 197:1457–1467
17. Vet NJ, de Wildt SN, Verlaat CW, et alA randomized controlled trial of daily sedation
interruption in critically ill children
. Intensive Care Med 2016; 42:233–244
18. Verlaat CW, Heesen GP, Vet NJ, et alRandomized controlled trial of daily interruption of sedatives in critically ill children
. Paediatr Anaesth 2014; 24:151–156
19. Gupta K, Gupta VK, Jayashree M, et alRandomized controlled trial of interrupted versus continuous sedative infusions in ventilated children
. Pediatr Crit Care Med 2012; 13:131–135
20. Grunwell JR, Travers C, McCracken CE, et alProcedural sedation
outside of the operating room using ketamine in 22,645 children
: A report from the Pediatric Sedation
Research Consortium. Pediatr Crit Care Med 2016; 17:1109–1116
21. Green SM, Roback MG, Krauss B, et alEmergency Department Ketamine Meta-Analysis Study Group: Predictors of emesis and recovery agitation with emergency department ketamine sedation
: An individual-patient data meta-analysis of 8,282 children
. Ann Emerg Med 2009; 54:171.e–80.e1
22. Grunwell JR, Travers C, Stormorken AG, et alPediatric procedural sedation
using the combination of ketamine and propofol outside of the emergency department: A report from the Pediatric Sedation
Research Consortium. Pediatr Crit Care Med 2017; 18:e356–e363
23. Kamat PP, Kudchadkar SRIV Clonidine in the PICU: Time for dexmedetomidine to share the limelight? Pediatr Crit Care Med 2018; 19:792–794
24. Hanning SM, Orlu Gul M, Toni I, et alCloSed Consortium: A mini-review of non-parenteral clonidine preparations for paediatric sedation
. J Pharm Pharmacol 2017; 69:398–405
25. Hayden JC, Breatnach C, Doherty DR, et alEfficacy of α2-agonists for sedation
in pediatric critical care: A systematic review. Pediatr Crit Care Med 2016; 17:e66–e75
26. Kleiber N, de Wildt SN, Cortina G, et alClonidine as a first-line sedative agent after neonatal cardiac surgery: Retrospective cohort study. Pediatr Crit Care Med 2016; 17:332–341
27. Kleiber N, van Rosmalen J, Tibboel D, et alHemodynamic tolerance to IV clonidine infusion in the PICU. Pediatr Crit Care Med 2018; 19:e409–e416
28. Sulton C, McCracken C, Simon HK, et alPediatric procedural sedation
using dexmedetomidine: A report from the Pediatric Sedation
Research Consortium. Hosp Pediatr 2016; 6:536–544
29. Gong M, Man Y, Fu QIncidence of bradycardia in pediatric patients receiving dexmedetomidine anesthesia: A meta-analysis. Int J Clin Pharm 2017; 39:139–147
30. Jun JH, Kim KN, Kim JY, et alThe effects of intranasal dexmedetomidine premedication in children
: A systematic review and meta-analysis. Can J Anaesth 2017; 64:947–961
31. Grant MJ, Schneider JB, Asaro LA, et alRandomized Evaluation of Sedation
Titration for Respiratory Failure Study Investigators: Dexmedetomidine use in critically ill children
with acute respiratory failure. Pediatr Crit Care Med 2016; 17:1131–1141
32. Venkatraman R, Hungerford JL, Hall MW, et alDexmedetomidine for sedation
during noninvasive ventilation in pediatric patients. Pediatr Crit Care Med 2017; 18:831–837
33. Shutes BL, Gee SW, Sargel CL, et alDexmedetomidine as single continuous sedative during noninvasive ventilation: Typical usage, hemodynamic effects, and withdrawal. Pediatr Crit Care Med 2018; 19:287–297
34. Cruickshank M, Henderson L, MacLennan G, et alAlpha-2 agonists for sedation
of mechanically ventilated adults in intensive care units
: A systematic review. Health Technol Assess 2016; 20:vxx, 1–117
35. Pan W, Wang Y, Lin L, et alOutcomes of dexmedetomidine treatment in pediatric patients undergoing congenital heart disease surgery: A meta-analysis. Paediatr Anaesth 2016; 26:239–248
36. Baxter MR, Bevan JC, Samuel J, et alPostoperative neuromuscular function in pediatric day-care patients. Anesth Analg 1991; 72:504–508
37. Kovac ALSugammadex: The first selective binding reversal agent for neuromuscular block. J Clin Anesth 2009; 21:444–453
38. Bom A, Bradley M, Cameron K, et alA novel concept of reversing neuromuscular block: Chemical encapsulation of rocuronium bromide by a cyclodextrin-based synthetic host. Angew Chem Int Ed Engl 2002; 41:266–270
39. Abrishami A, Ho J, Wong J, et alSugammadex, a selective reversal medication for preventing postoperative residual neuromuscular blockade. Cochrane Database Syst Rev 2009; (4):CD007362
40. Abrishami A, Ho J, Wong J, et alCochrane corner: Sugammadex, a selective reversal medication for preventing postoperative residual neuromuscular blockade. Anesth Analg 2010; 110:1239
41. Sanchez-Pinto LN, Nelson LP, Lieu P, et alImplementation of a risk-stratified opioid weaning protocol in a pediatric intensive care unit. J Crit Care 2018; 43:214–219
42. Robertson RC, Darsey E, Fortenberry JD, et alEvaluation of an opiate-weaning protocol using methadone in pediatric intensive care unit patients. Pediatr Crit Care Med 2000; 1:119–123
43. Dahl RESleep and the developing brain. Sleep
44. Everson CASustained sleep
deprivation impairs host defense. Am J Physiol 1993; 265:R1148–R1154
45. Kudchadkar SR, Aljohani OA, Punjabi NMSleep of critically ill children
in the pediatric intensive care unit: A systematic review. Sleep
Med Rev 2014; 18:103–110
46. Kudchadkar SR, Yaster M, Punjabi AN, et alTemporal characteristics of the sleep
EEG power spectrum in critically ill children
. J Clin Sleep
Med 2015; 11:1449–1454
47. Kudchadkar SR, Beers MC, Ascenzi JA, et alNurses’ perceptions of pediatric intensive care unit environment and work experience after transition to single-patient rooms. Am J Crit Care 2016; 25:e98–e107
48. Traube C, Silver G, Reeder RW, et alDelirium in critically ill children
: An International Point Prevalence Study. Crit Care Med 2017; 45:584–590
49. Patel AK, Biagas KV, Clark EC, et alDelirium in the pediatric cardiac extracorporeal membrane oxygenation patient population: A case series. Pediatr Crit Care Med 2017; 18:e621–e624
50. Patel AK, Biagas KV, Clarke EC, et alDelirium in children
after cardiac bypass surgery. Pediatr Crit Care Med 2017; 18:165–171
51. Traube C, Silver G, Gerber LM, et alDelirium and mortality in critically ill children
: Epidemiology and outcomes of pediatric delirium
. Crit Care Med 2017; 45:891–898
52. Traube C, Mauer EA, Gerber LM, et alCost associated with pediatric delirium
in the ICU. Crit Care Med 2016; 44:e1175–e1179
53. Smith HA, Boyd J, Fuchs DC, et alDiagnosing delirium
in critically ill children
: Validity and reliability of the Pediatric Confusion Assessment Method for the intensive care unit. Crit Care Med 2011; 39:150–157
54. Luetz A, Gensel D, Müller J, et alValidity of different delirium
assessment tools for critically ill children
: Covariates matter. Crit Care Med 2016; 44:2060–2069
55. Smith HA, Gangopadhyay M, Goben CM, et alThe Preschool Confusion Assessment Method for the ICU: Valid and reliable delirium
monitoring for critically ill infants and children
. Crit Care Med 2016; 44:592–600
56. Simone S, Edwards S, Lardieri A, et alImplementation of an ICU bundle: An interprofessional quality improvement project to enhance delirium
management and monitor delirium
prevalence in a single PICU. Pediatr Crit Care Med 2017; 18:531–540
57. Mody K, Kaur S, Mauer EA, et alBenzodiazepines and development of delirium
in critically ill children
: Estimating the causal effect. Crit Care Med 2018; 46:1486–1491
58. Smith HAB, Gangopadhyay M, Goben CM, et alDelirium and benzodiazepines associated with prolonged ICU stay in critically ill infants and young children
. Crit Care Med 2017; 45:1427–1435
59. Meyburg J, Dill ML, Traube C, et alPatterns of postoperative delirium
. Pediatr Crit Care Med 2016; 1:128–133
60. Barnes SS, Grados MA, Kudchadkar SRChild psychiatry engagement in the management of delirium
in critically ill children
. Crit Care Res Pract 2018; 2018:9135618
61. Slooff VD, van den Dungen DK, van Beusekom BS, et alMonitoring haloperidol plasma concentration and associated adverse events in critically ill children
: First results of a clinical protocol aimed to monitor efficacy and safety. Pediatr Crit Care Med 2018; 19:e112–e119