Fig. 27.1
Maslow’s hierarchy of needs (1943)
Furthermore, both the growing concern on potential long-term neurological adverse effects of prolonged sedation as well as the modern trend to minimize (or interrupt regularly) sedation, to avoid longstanding immobilization and to use new sedatives with a clearly lighter sedation profile (e.g., dexmedetomidine) result in a growing interest in NPS. Finally, some preliminary evidence suggests that the reduction of environmental stimuli and stress by use of NPS may improve healing and decrease the need for sedation [12, 13].
A recent qualitative survey among family members and nurses on an adult ICU showed that the top four non-pharmacological interventions found to be useful, relevant, and feasible were music therapy and distraction (cognitive-behavioral category), simple massage (physical category), and family presence facilitation (emotional support category) [14]. The potential advantages of healing environmental interventions have been described in an adult ICU [15]. Currently, no similar research on PICUs is available. Despite this limited evidence we believe that at least four relevant strategies can be identified as essential for a policy directed towards optimal comfort experience. Definitely more research is needed to explore the effects of these strategies on patient comfort.
Careful Identification and Treatment of Underlying Causes of Anxiety, Agitation, and Insomnia
Distress in pediatric patients is rarely a spontaneous behavior but rather the result of an underlying problem. At first, potentially (life) threatening causes need to be excluded, such as hypoxemia, imminent respiratory or circulatory failure, and hypoglycemia. Also delirium and (sub-optimally managed) pain need to be excluded systematically. “Fighting-the-ventilator” can often be reduced by carefully tailoring a support ventilatory mode (including optimal trigger-function, inspiratory times, and airway pressures) according to the individual child’s respiratory needs. Finally hunger, cold, bladder retention, constipation, malpositioning (e.g., infants may prefer a prone position), fear for medical procedures and lack of age-appropriate comfort may all contribute to distress.
Child Centered Care
Verbal and non-verbal communication should be adapted to the child’s (developmental) age. Whereas young infants may benefit from gentle and tactile interaction while reducing exposure to ambient stimuli (e.g., by swaddling) within a warm “shelter,” older children may prefer more control and sufficient sight on toys, television, or tablets. Guided imagery, storytelling, age-adapted music, presence of parents and siblings at any time and even animal-assisted therapy are all reasonable interventions to optimize feelings of comfort and normalize the ICU experience [16].
Family-Centered Care
Family-centered care (FCC) is increasingly considered to be a key component of high quality health care for children and their parents/caretakers. FCC can be characterized by health care practices that are respectful of and responsive to individual patient and family preferences, needs, and values. In FCC a major role of health care providers is to enable and empower patients and their family members to play an active role in healthcare decision-making and treatment. When done well and consistently, FCC can lead to safer, more personalized and effective care, improved health care experiences and patient outcomes, and more responsive organizations [17]. Although no research has been published on the effects of FCC on children’s experiences during their stay on a PICU , it is very likely that an FCC approach will result in patient comfort management that better meets with individual needs. Furthermore, involving parents and siblings in daily care may result in more confidence and collaboration.
Environmental Control
Traditionally, ICUs are process-focused departments designed and constructed to proceed efficiently within a setting of high workload and complex technical support. Lightning conditions are usually bright and uninterrupted in order to optimize patient observation and interventions. Furthermore, sensitive alarm settings of all kind of technical devices result in many auditory stimuli, not to mention the inter-professional communication as a substantial source of noise. Excellent evidence exists on the negative effects of this type of busy environment on feelings of comfort, incidence of delirium and agitation, and need for sedation [15, 18].
In a well-intended attempt to camouflage the medical atmosphere, child friendly gadgets, bright colors, cartoons, comic characters or art may be added to the interior design of patient rooms. Although no systematic research whatsoever exists within this specific field, there are several reasons to question the effectiveness of this common practice. The most important criticism to this design lies in its usually static character that limits individualized modifications according to the (developmental) age and individual characteristics or needs of a child. Comic characters that are popular in one age group or gender may be meaningless or even frightening or childish in another. It has been shown that children’s emotional associations with colors are determined by age and gender [19]. Furthermore, a recent study on the stress-reducing effects of art in pediatric health care indicated that children may be less in tune with art and possibly more affected by social support, e.g. parental care [20]. In a recent qualitative study, parents of children with autism spectrum disorder indicated that “autism-friendly” modifications to the built environment (being: quiet and/or sensory rooms, warm colors, no fluorescent lights) would result in less stressful visits [21].
In the last decade there is a growing interest in the effects of the physical, built environment on patient safety, caring processes and staff efficiency but also on the general well-being and healing in patients [22, 23]. This has resulted in fascinating new design objectives for patient facilities [24], including the creation of highly innovative healing environments for children’s hospitals and pediatric units [25, 26]. Design considerations for children’s health and emergency services have been published and include family-centered approach, separation from adult patients, privacy and acoustic control, a child and family friendly environment and provision of play and entertainment [27]. A quieter environment, one that includes familiar persons, dotted with windows and natural light, creates a space that makes people feel balanced and reassured.
Clinical Pharmacology of Intravenous Analgesics and Sedatives
In this section the most important drugs currently used for analgesia and sedation in children (1 month–18 years old) admitted to the PICU will be discussed. Despite their widespread use evidence-based dosages of most drugs are scarce. Furthermore these dosages are usually not supported by pharmacokinetic or pharmacodynamics models. Applying such models could enhance tailoring of dosages to the individual patient and prevent for instance both under- and over-dosing [28]. Table 27.1 summarizes the most commonly used analgo-sedative drugs with their doses used in a PICU .
Table 27.1
Loading and maintenance dosages for the most import iv drugs currently used for analgesia and sedation in critically ill children
Loading dose | Maintenance dose | T½ | Dependency and withdrawal | ||
|---|---|---|---|---|---|
Morphine | 50–100 μg/kg | 5–40 (−100) μg/kg/h | 2–3 h | ++ | ++ |
Fentanyl | 0.5–2 μg/kg | 0.5–5 μg/kg/h | 3–4 h | ++ | ++ |
Remifentanil | 0.5–1 μg/kg | 0.1–2 μg/kg/min | 3–10 min | ++ | ++ |
Midazolam | 50–100 μg/kg | 50–400 (−1000) μg/kg/h | 3–11 h | ++ | ++ |
Lorazepam | 20–40 μg/kg | 20–100 μg/kg/h | 8–15 h | ++ | ++ |
Dexmedetomidine | 0.05–1 μg/kg (10–20 min) | 0.05–0.7 (−2) μg/kg/h | 2–2 ½ h | +/− | + |
Clonidine | 1–3 μg/kg | 0.25–3 μg/kg/h | 5–25 h | +/− | +/− |
Propofol | 1–3 mg/kg | 1–4 mg/kg/h (max 48 h) | 30–60 min | − | − |
Ketamine | 1–2 mg/kg | 5–60 μg/kg/min | 2–4 h | − | − |
Opioids
Pain is one of the key factors of discomfort and agitation in critically ill children. Opioids are therefore one of the most prescribed drugs in the pediatric ICU. There are several opioids currently in use. All opioids have mainly analgesic effects, relieving visceral, somatic, and neuropathic pain. In higher dosages they also offer sedation and anxiolysis. The analgesic effect is mainly the result of binding to μ-receptors and k-receptors, and (for some synthetic opioids) to δ-receptors. It is thought that the main side effects are caused through binding to other receptors [29].
Several natural and synthetic opioids are available for clinical use in pediatric patients. Morphine is the only natural opioid currently used and fentanyl and remifentanil are the most used synthetic opioids in this population. Sufentanil is mainly used outside the ICU in operating rooms. Other opioids such as codeine, oxycodone, methadone, and meperidine are usually used for analgesia outside the ICU and therefore not included in this overview. Morphine and fentanyl are currently the most used opioid analgesics in critically ill children [30, 31]. Morphine may offer advantages over fentanyl because it induces less tolerance in general and causes less withdrawal symptoms in neonates [32, 33]. As far as we know no studies are available comparing morphine with fentanyl for analgesia in critically ill children.
The main advantage of opioids is the strong analgesic effects and the relatively large therapeutic window. Given their relatively limited sedative effects, opioids are rarely indicated when pain can reasonably be excluded.
Opioids can be administered either by intermittent boluses or by continuous infusion. In children with severe vaso-occlusive disease, continuous infusion of morphine provided better analgesia than intermittent opioid therapy [34]. A randomized double-blind clinical trial comparing the efficacy of 10 mcg/kg/h morphine (continuous intravenous infusion) with that of 30 mcg/kg morphine every 3 h (bolus intravenous injection) in 181 young children aged 0–3 years, admitted on a PICU following major abdominal or thoracic surgery showed no differences in reducing postoperative pain. However, in children >1 year old continuous infusion turned out to be favorable [35]. Adding paracetamol and/or nonsteroidal anti-inflammatory drugs (NSAIDs) to the pain treatment reduces opioid requirements and may also reduce adverse effects [36]. Therefore the use of opioids should always be accompanied by either paracetamol or an NSAID.
Side Effects
The most important side effect of all opioids is respiratory depression, caused by direct inhibitory effects on the brainstem respiratory centers. This effect is dose-dependent and may occur before consciousness is impaired. The decrease in minute volume is primarily due to a decreased respiratory rate, while tidal volume remains usually intact. Opioids also cause a diminished sense and reaction to CO2 levels. As a consequence, fatal accidents related to morphine toxicity use are nearly always due to respiratory arrest. Newborn infants, patients with pre-existing respiratory problems (e.g., chronic respiratory failure) and recurrent hypoxic events (e.g., obstructive sleep apnea syndrome) have a higher risk of opioid associated respiratory depression[37]. In critically ill children on mechanical ventilation, however, the respiratory depressive effects may in fact be advantageous, as they may enhance patient-ventilator synchrony. Morphine can also cause hypotension, which is especially important in hemodynamically instable or cardiac patients.
Opioids are also particularly known for their gastro-intestinal side effects such as the prolongation of gastric emptying time and the reduction of pro-pulsatile activity in the small and large intestines. These mechanisms explain the most important gastro-intestinal adverse effects. Nausea and vomiting may be associated with the use of any opioid, although it is mostly seen in ambulatory patients while it seems to be only a minor problem in ventilated (not-moving) patients. A far more important problem for the PICU patient is constipation, which occurs in 40–95 % of patients treated with opioids. The best way to treat this is by starting laxatives early in the course of opioid administration. Enemas may be an effective alternative .
Pruritus is another common problem with opioids, especially with morphine, causing potential additional discomfort. Since this effect is at least partially caused by histamine release anti-histamines may be a reasonable choice [29]. Also opioid-induced disinhibition of itch-specific neurons in the spinal dorsal horn may play a role [38]. Lowering the dose is usually an effective strategy. Sometimes the pruritus leads to agitation in such extent that switching to a synthetic opioid seems to be the best choice [39].
Another common problem with especially continuous administration of opioids is the occurrence of a retention bladder caused by an inhibitory effect on voiding reflexes and increasing the external sphincter tone with resultant increase in bladder volume [29]. Especially in young children this can be a problem because they cannot properly convey where they feel pain, which can cause a delay in treating a retention bladder. When thought of, it is easily resolved by inserting a urinary catheter.
Tolerance and withdrawal are also very common with opioids. Dependency can even occur after only a few days of opioid use [29, 39]. A large prospective study has shown that prolonged opioid exposure leads to a significant increase in dose. An opioid exposure of 7 or 14 days, for instance, requires doubling of the daily opioid dose in 16 and 20 % of patients, respectively. This doubling occurred in 43 % of those receiving opioids for 14 days or more. Doubling of the opioid dose was furthermore more likely to occur in the case of co-therapy with benzodiazepines and less likely to occur if morphine was used as the primary opioid [32]. Besides a lower tendency to deliver tolerance morphine also has a significantly lower prevalence of developing withdrawal symptoms in neonates compared to fentanyl [33].
Withdrawal symptoms after cessation can consist of a hyper alert state, hyperalgesia, hyperthermia, and diarrhea [29]. Withdrawal symptoms can be prevented by gradually tapering the intravenous dose or by switching to a long-acting oral opioid (e.g., methadone) [39]. This will be further discussed in the section withdrawal in this chapter.
Finally, neuro-apoptosis has come under growing attention as a possible relevant adverse effect of opioids. Recent animal studies have shown that, especially in the developing brain, neuro-apoptosis occurs with long-term opioid infusion [40–42]. Evidence from research in humans remains limited, however. A follow-up study of critically ill neonates treated with continuously administered morphine infusion of 10 μg/kg/h demonstrated, however, that it didn’t harm general functioning and may even have a positive influence on executive functions at the age of 8–9 years [43]. In children with meningococcal septic shock the use of opioids during admission was associated with long-term adverse neuropsychological outcome independent of severity of illness scores. These results are limited, however, by the fact that it concerns a retrospective cohort study [44].
Morphine
Morphine is a water-soluble molecule whose low-lipid-solubility accounts for its slow penetrance into the brain causing its delayed onset of clinical effect [29]. Peak effect occurs 20 min after administration. The duration of action is about 2 h following a single iv loading dose of 100 μg/kg. Maintenance dosages up to 60 μg/kg/h are safe to use but usually a dose within the range of 10–30 μg/kg/h is sufficient in most postoperative patients with pain [45].
Morphine undergoes extensive hepatic and extra-hepatic glucuronidation. Metabolites are excreted primarily in the urine. Only one of the metabolites of morphine is active, namely morphine-6-glucuronide. Accumulation of this active metabolite in renal disease may cause prolonged opioid effects. Neonates have a slower metabolization and excretion and it is thought that they form more morphine-3-glucuronide. This metabolite is thought to cause anti-analgesic effects. Towards the age of one month metabolization is similar compared to adults [45]. It is well known that glomerular filtration rate and tubular secretion are reduced in neonates and therefore dose reduction might be necessary to prevent excessive accumulation. This of course is also true for patients with renal impairment in general.
Fentanyl
Fentanyl is one of the most currently used synthetic opioids. It is the second most used analgesic after morphine and is about 100 times more potent compared to morphine. It has a relatively short half-life of about 3–4 h. Especially in neonates it is the drug of first choice when IV analgesia is needed. This is mainly due to the difference in metabolization compared to morphine. Fentanyl undergoes hepatic metabolization and is then renally excreted, and it has no known active metabolites. A disadvantage of fentanyl is the fact that it is stored in peripheral tissue, which can lead to accumulation with repeated doses and continuous infusion. This can lead to a prolonged half-life up to 21 h.
Because of the relatively short half-life time a bolus is not necessary in every increase in maintenance dose. But boluses are often used before handling the patient or before painful interventions. Bolus dosages of 1–2 μg/kg/h are usually sufficient. Fentanyl is usually dosed at 0.5–5 μg/kg/h, but dosages up to 10 μg/kg/h are proven to be safe [45]. One of the main complications, both in non-ventilated and in ventilated patients is chest wall rigidity, especially after fast infusion of a bolus and in younger children especially neonates [39]. Therefore it is recommended to give boluses by slow infusion up to 10 min. Fentanyl associated acute chest wall rigidity can be successfully treated with instant antagonizing drugs (naloxone (10–40 μg/kg IV) or with direct administration of a neuromuscular blocking agent.
Remifentanil
Remifentanil is a synthetic opioid that is relatively new, but upcoming in pediatric ICUs. It is a strong opioid agonist, and it is comparable to fentanyl in potency but ultra-short acting with an onset time of 1–3 min. It has a half-life time of 3–10 min, which makes it ideal for weaning a patient from mechanical ventilation. Remifentanil is usually dosed at 0.1–2 μg/kg/min. Most experience is with short-term, procedure-related analgesia, which is usually achieved by repeated administration of small bolus dosages (up to 1 μg/kg) up to a desired analgesic effect is reached [46]. There is very limited evidence on the use of remifentanil as a continuous IV infusion for analgesia in critically ill children [47].
Remifentanil is metabolized by plasma esterases into minimally active metabolites resulting in a very short duration of action (even after repeated or prolonged administration) and an elimination that is independent of hepatic metabolism or renal excretion.
Conclusions
Opioids are drugs with strong analgesic effects making them very useful for analgesia in critically ill children but should be accompanied by either paracetamol or an NSAID to reduce the dose of the opioid. The side effects are mostly preventable or easily treatable. Therefore it is a good choice as the primary drug in critically ill children to provide analgesia when (severe) pain is involved. Morphine should be used with some care in neonates because of different metabolization and fentanyl seems to be the better choice in these patients. With all opioids there is a high risk of dependency and withdrawal after cessation, and therefore slow tapering or converting to oral administration of methadone is necessary. There is some evidence that opioids cause neuro-apoptosis in long-term sedation, especially in the developing brain. At this moment evidence is still conflicting and therefore there is no reason to change current practice.
Benzodiazepines
Benzodiazepines are the most used sedatives in critically ill children. Midazolam and secondly lorazepam are the preferred benzodiazepines for sedation in mechanically ventilated children accounting for, respectively, 82–97 % and 1–16 % of used benzodiazepines. Diazepam is very rarely used for sedation in this patient category [48]. No studies in the pediatric population directly compared lorazepam and midazolam for sedation in the intensive care unit.
Benzodiazepines mainly bind with χ-aminobutyric acid A receptor (GABA-A receptors). GABA is the most important inhibitory neurotransmitter in the CNS. Binding to the GABA-receptors causes a hyperpolarization of neurons, leading to a decreased sensitivity to excitatory neurotransmitters and, consequently, to CNS depression. The clinical effects of benzodiazepines on the central nervous system are dose-dependent, resulting in a progressive increase of sedative effects: anxiolysis, moderate sedation, deep sedation, full loss of consciousness (coma), and eventually death [29, 49]. Next to sedative effects, benzodiazepines have also amnesic and anti-epileptic properties.
One of the advantages of benzodiazepines compared to many other sedatives such as propofol or dexmedetomidine is that it provides a pronounced stability because of long plasma half-life time. In long-term sedation it can provide stability making the child less at risk for accidental extubation and catheter removal [50, 51]. On the other hand, the long plasma half-life time may impede short-term awakening that may be desirable in particular patient categories (e.g., patients with traumatic brain injury and need for repeated neurological assessment; postoperative patients with expected early extubation) [52].
Midazolam is metabolized by cytochrome P450 (CYP) enzymes and by glucuronide conjugation whereas lorazepam directly undergoes glucuronide conjugation. Midazolam and lorazepam thus have different routes of metabolization [53]. This difference may be relevant when CYP450 activity could be altered due to pharmaco-genetic variability, disease conditions or interactive co-medication. Since the metabolites of midazolam are also active sedative agents, impaired excretion may lead to oversedation, especially in long-term sedation and renal failure [54]. Other drugs which can inhibit CYP enzymes (e.g., macrolide antibiotics, certain antiviral drugs, antifungal drugs (itraconazole, ketoconazole) and grapefruit juice) can cause significant interactions leading to stronger effects of midazolam due to its reduced clearance. It has also been shown that increased disease severity reduces midazolam clearance in critically ill children by inflammation mediated CYP3A downregulation and by organ failure in itself [55].
Side Effects
The most important side effects of benzodiazepines in children are respiratory depression and apnea, which will be aggravated when combined with opioid analgesics. Furthermore benzodiazepines may cause vasodilatation leading to hypotension, especially in neonates and after fast infusion of midazolam [29, 31]. These effects may be more pronounced in hemodynamically instable patients or hypovolemic patients. Both midazolam and lorazepam are also associated with mild gastro-intestinal side effects such as nausea and vomiting [29]. Some children are known to react paradoxically to midazolam with agitation and restlessness needing conversion to other sedatives [56, 57].
Although benzodiazepines are efficient in providing sedation, anxiolysis and amnesia they carry several important disadvantages. In adults the use of benzodiazepines has been associated with an increased risk of development of delirium [58, 59]. Benzodiazepines also impose a high risk for the development of withdrawal symptoms especially after prolonged administration and/or use of high dosages [60, 61]. Some researchers claim that withdrawal symptoms can occur even after only 3 days of use [61]. Abrupt cessation is therefore not recommended. In case of necessity of abrupt cessation it is possible to switch to oral lorazepam, to prevent withdrawal from both lorazepam and midazolam [62]. Withdrawal will be further discussed in another section of this chapter.
In adults it has been shown that the use of benzodiazepines for ICU sedation is associated with a prolonged length of hospital stay and longer duration of mechanical ventilation compared to non-benzodiazepine sedation regimens [63].
Midazolam
Midazolam is more lipid soluble compared to lorazepam, resulting in a quicker onset of sedation and a larger distribution volume [54]. It has a slightly shorter half-life time compared to lorazepam. Elimination half-life time is 3–11 h, depending on individual clearance characteristics and duration of administration [54].
In most children dosages of 50–300 μg/kg/h are sufficient to reach adequate sedation. There is some evidence that it might not even be beneficial to administer higher doses than 300 μg/kg/h for the purpose of sedation [65, 66]. In patients with status epilepticus dosages up to 1600 μg/kg/h are used for a limited duration of time [67]. Because of long plasma half-life time a bolus of 50–100 μg/kg should be given whenever maintenance doses are increased.
Lorazepam
Lorazepam is twice as potent compared to midazolam [68]. As mentioned before it has a slightly longer plasma half-life time. Elimination half-life time is 8–15 h [54]. In most children dosages of 20–100 μg/kg/h are sufficient. Because of the long plasma half-life time a bolus of 20–40 μg/kg is recommended in order to reach a steady state [39].
Conclusions
Benzodiazepines are suitable drugs for sedation with strong sedative effects. As they have no analgesic effects whatsoever they should always be combined with adequate analgesia whenever pain is involved. Because the plasma half-life time may substantially increase following continuous administration benzodiazepines may not be the best choice in patients who are expected to only need sedation for a short period of time. Benzodiazepine use is an independent predictor of prolonged mechanical ventilation and length of hospital stay.
NMDA-Receptor Antagonist: Ketamine
Ketamine is a phencyclidine derivative and its main action is antagonizing NMDA-receptors in the brain. It has sedative, dissociative, and analgesic effects. Ketamine is a racemic mixture of an active (S) and a (practically) inactive (R) enantiomer. The R-enantiomer, however, interferes with the clearance of the S-enantiomer. The S-isomer alone (available for clinical use in some European countries as S-Ketamine or Esketamine) has less side effects, a faster recovery and more or less twice the potency compared to the racemic compound. Dosages of S-ketamine are therefore about 50 % of those published for racemic Ketamine.
Because of the high hydrophobic chemical characteristics ketamine preferentially moves into highly perfused and lipophilic tissues such as the brain and spinal cord and therefore has a large volume of distribution. This causes the rapid onset of action, even after a single loading dose. Subsequently, blood levels fall rapidly, resulting in drug redistribution out of the CNS back into the blood and other tissues. This accounts for the short time of action, and this occurs a lot faster than metabolization [29].
Metabolization occurs in the liver by forming norketamine, which has reduced CNS activity. Norketamine is then further metabolized and excreted in urine and bile. Ketamine is rapidly cleared making it suitable for continuous infusion.
Clinical Uses
There is hardly any evidence supporting the use of ketamine for prolonged sedation in pediatric ICU patients. However, it has been used in doses ranging from 5 to 60 μg/kg/min for the sedation of children who respond poorly to the opioid-benzodiazepine combination [69, 70].
Because of its secondary sympathomimetic, bronchodilating effects, Ketamine has been recommended for use in children presenting with severe asthma [71]. A limited number of case reports have reported beneficial effects on bronchoconstriction following ketamine administration, including non-intubated children [72]. A recent Cochrane review, however, showed the limited evidence for routine use in severe asthma and the absent evidence for a beneficial effect in intubated asthma patients [73].
More recently, low-dose peri-operative ketamine administration has been shown in a meta-analysis to decrease postoperative pain intensity and non-opioid analgesic requirement. However, ketamine failed to exhibit a postoperative opioid-sparing effect [74].
One of the main advantages of ketamine over many other sedatives is its little effect on the respiratory drive. Induction doses of ketamine might have a minor effect on respiratory minute volume, but the risk of respiratory depression or apnea is much less compared to many other sedatives. Furthermore retention of protective pharyngeal and laryngeal reflexes makes it a popular drug for procedural analgosedation.
Unlike many other sedatives, ketamine typically increases hearth rate, mean arterial pressure and cardiac output. These effects are most likely mediated by inhibition of catecholamine reuptake. Therefore it seems to be a good choice in patients at risk for hypotension [70]. However, in patients with pronounced circulatory failure and thus relatively depleted of catecholamines (e.g., septic shock) ketamine may cause deep hypotension and cardiac arrest [75].
Side Effects
One of the main reasons why ketamine is not much used for sedation in critically ill patients is the risk of hallucinations and delirium. These side effects seem to occur less frequently in children (about 10 % of cases), although no systematic research has been done within this perspective. There is some limited evidence, but no proof, that these effects are counteracted by simultaneous use of a benzodiazepine [39, 76].
Furthermore ketamine increases salivation and mucus production, which can be a very unwanted effect in mechanically ventilated patients .
It is often mentioned that ketamine causes increased intracranial pressure thereby making it unusable in traumatic brain injury. A recent systematic review, however, suggests that ketamine does not adversely affect intracranial or cerebral perfusion pressures compared with other intravenous agents. This study, however, only studied ketamine as an inductive agent and is further limited by the lack of large, randomized trials addressing this topic [77].
Ketamine also seems to cause neuro-apoptosis in long-term use [64]. As with many other sedatives there is no clear evidence of the neurodevelopmental consequences.
Conclusions
Ketamine is a strong sedative and analgesic drug that can be used for continuous sedation in critically ill children. It is a very attractive drug in short-term sedation and procedural sedation because it does not cause respiratory depression and has a very short half-life time. It also seems to be a safe choice in patients with cardiovascular instability but should not be used in patients with pronounced circulatory failure. The main side effects of delirium and hallucinations seem to be of less importance in children compared to adults, but it does occur in children as well which can be a reason to discontinue ketamine or add a benzodiazepine to the sedation regiment.
Alpha-2-Receptor Agonists
Alpha-2-receptor agonists are among the newer used sedatives in pediatric ICUs. They have a range of effects including sedation, analgesia, and anxiolysis. Clonidine has been used for hypertension for many years and was discovered to also be effective for withdrawal symptoms. Later on its sedative properties became more apparent. Dexmedetomidine is more recently discovered as a highly selective alpha-2-receptor agonist with even less side effects compared to clonidine . The selectivity ratio of alpha-2 compared to alpha-1 stimulation is 1620:1 for dexmedetomidine compared to 220:1 for clonidine [78]. The main sedative actions of both agents are achieved by stimulation of alpha-2-receptors in the central nervous system, especially in the locus ceruleus and the dorsal horn of the spinal cord [79]. Currently, dexmedetomidine is not yet approved for prolonged use in pediatric patients. Nevertheless, there is growing evidence that long-term use for sedation in pediatric ICU patients is safe and effective [79, 80].
One of the typical characteristics of alpha-2-receptor agonists is that they generate sedation without limiting patient interaction. Particularly for dexmedetomidine this typical effect has been described as “generating natural sleep” [81, 82]. This means that all non-pharmacological means of providing comfort are even more important to reach effective sedation with dexmedetomidine alone. Compared to Propofol and midazolam, the use of dexmedetomidine has been shown to maintain patient interactivity, resulting in a better opportunity for patients to communicate their needs. Alpha-2-receptor agonists may be less suitable when a painful or otherwise unpleasant stimulus is present [80].
A particular advantage of alpha-2 receptor agonists is the absence of respiratory depressant effects [83]. The combination of these advantages over many other sedatives are thought to be the explanation for the reduction in days of mechanical ventilation and length of hospital stay, proven in adults when alpha-2-agonists are compared to benzodiazepines or propofol [83–85]. Also in children the use of dexmedetomidine is associated with a shorter duration of mechanical ventilation when compared with the use of fentanyl in postoperative pediatric cardiac surgical patients [86]. Besides all of these advantages, alpha-2 receptor agonists seem to give a decreased risk of both delirium and oversedation, compared to traditional sedatives [83].
Elimination is through both hepatic metabolism to inactive metabolites and direct renal excretion [29, 87].
Alpha-2-agonists provide analgesia and sedation and are therefore a good alternative to other sedatives but can also be used as an adjuvant especially in addition or replacement of benzodiazepines [79, 88]. Studies directly comparing dexmedetomidine to midazolam are rare. The prospective study by Tobias et al. shows that at a dose of 0.25 μgr/kg/h dexmedetomidine was approximately equivalent to midazolam at 0.22 mg/kg/h. At higher infusion rates of 0.5 μgr/kg/h it provides more effective sedation as demonstrated by a need for fewer doses of morphine as well as better sedation scores [89]. A retrospective study by Whalen et al. didn’t notice a significant difference in the dosing of opiates or benzodiazepines after starting dexmedetomidine. This study is, however, biased by the retrospective nature of it [79]. For clonidine it has been shown that it can be used as an alternative for midazolam as a primary sedative agent [87, 90]. In ventilated newborns clonidine reduced fentanyl and midazolam demand with deeper levels of analgesia and sedation without substantial side effects. This effect was not seen in older children [91]. More and larger randomized controlled trials investigating the additive effect of dexmedetomidine and clonidine and/or the direct comparison with benzodiazepines are therefore urgently needed. Also studies comparing clonidine with dexmedetomidine are lacking.
Therefore, the exact place of these drugs for sedation of critically ill children, including their optimal and maximum dosages, still has to be determined.
Side Effects
Alpha-2-agonists are, like other sedatives, associated with tolerance and, consequently, withdrawal following discontinuation [79, 92]. The main side effects of alpha-2-agonists are, in accordance with stimulating Θ-receptors, of cardiovascular origin. The major side effects therefore are bradycardia and hypotension. Because of the higher alpha-2-selectivity, dexmedetomidine has less cardiovascular effects than clonidine .
Hypertension is described as well, which is thought to be due to alpha-1 agonistic and peripheral alpha-2B agonistic effects, leading to peripheral vasoconstriction [93]. This hypertension is described after intravenous start of clonidine , which then subdues shortly after [29]. It is also described when higher doses of dexmedetomidine are used. Lowering the dose may be indicated to reverse this effect. Dexmedetomidine administration has been shown to be safe in children with (congenital) heart disease. In this category of patients its use has been associated with decrease of opioid and benzodiazepine requirement and also a decreased need for inotropic support [94]. Similar findings have been observed in children with heart failure [95]. However, we believe that alpha-2-agonists need to be used cautiously in patients at risk for bradycardia, hypotension , and hypertension and that bolus dosages should be avoided especially in the latter patient categories.
Clonidine
Intravenous dosages for clonidine range from 0.25 up to 3 mcg/kg/h [83, 87]. Some studies describe a loading dose of 1–3 μg/kg [83, 87]. Others state that a loading dose is not indicated as continuous infusion will achieve sedation goals quickly and loading dose may provoke hypotension and/or bradycardia.
Oral clonidine is used for withdrawal symptoms after long-term use of opioids, benzodiazepines as well as alpha-2-receptor agonists. In addition, clonidine has independent local effects on tubular function which promote both diuresis and natriuresis (http://www.journalslibrary.nihr.ac.uk/hta/volume-18/back-section-2.html-ref1-bib37). In adults the peripheral α1 effects can cause hypertension and vasoconstriction in overdose, but this appears to be far less common in children [87].
Dexmedetomidine
Dexmedetomidine is a new generation alpha-2-receptor agonist with a high Θ2-selectivity. It has potent sedative and modest analgesic properties.
Usually dexmedetomidine is administered intravenously at a dose ranging from 0.2 to 0.7 μg/kg/h. Several studies have demonstrated that dosages up to 2 μg/kg/h are safe and sometimes necessary to achieve sufficient sedation scores [96]. Higher dosages may lead to hypertension as one case report suggests [97]. Some studies propose a slow-loading dose of 0.05–1 μg/kg in 10–20 min to achieve sedation earlier [79, 96]. As with clonidine giving a loading dose may provoke hypotension and/or bradycardia.
Conclusions
Alpha-2-agonists are a relatively novel group of sedatives, but with great potency. They provide sedation combined with moderate analgesia. They can be used as a sole sedative or as an adjuvant, which may lead to lower dosages of the primary used sedative. The most relevant side effects are bradycardia and hypotension, and therefore they should be used with care in hemodynamically unstable patients.
Propofol
Propofol is a short-acting, lipophilic intravenous anesthetic. Propofol causes global CNS depression, presumably through agonism of GABAA receptors and perhaps reduced glutamatergic activity through NMDA-receptor blockade. Propofol has no analgesic effects. Propofol is metabolized in the liver by conjugation to sulfate and glucuronide and further metabolized to less active metabolites that are renally excreted. Since hepatic clearance exceeds hepatic blood flow, biological availability following oral ingestion is zero. Clearance is decreased in neonates with a high risk of accumulation and therefore lower doses are advised.
One advantage of propofol over many other sedatives is the low risk of dependency and tolerance. Therefore propofol does not need to be weaned off slowly.
Side Effects
Although propofol is one of the most administered sedatives in adult ICUs, its clinical use in children is limited. This is because of the risk of propofol infusion syndrome (PRIS) , which is more common in children compared to adults. PRIS is a potentially life threatening complication, seen with prolonged and/or high dose propofol infusion. It’s characterized by metabolic acidosis, hyperlipidemia, rhabdomyolysis, and eventually bradyarrhythmias and asystole. Many case reports describe the fatal ends in children with PRIS, and only a few reports on survivors are found [101]. Especially young children, children with head injury and children with an underlying mitochondrial dysfunction seem to have a high risk for PRIS. Especially the characteristically irreversible fatal course once PRIS has become symptomatic makes that pediatric ICUs are rather reluctant using propofol for long-term sedation. If used, dosages of 1–4 mg/kg/h seem to be sufficient for sedation in critically ill children. For anesthesia dosages up to 18 mg/kg/h are proven to be safe. Usually a loading dose of 1–3 mg/kg is used to reach adequate sedation within minutes.
Most guidelines recommend infusion for a maximum of 48 h and a maximum of 4 mg/kg/h to avoid PRIS [102]. Appropriate monitoring by frequently measuring lactate and creatine kinase levels is still warranted under these circumstances [103].
One of the major common side effects of propofol is hypotension, which is dose-dependent and caused by vasodilation and possibly mild depression of myocardial contractility. Therefore propofol should be used with caution in hemodynamically instable patients. Furthermore propofol has a risk of respiratory depression, a side effect which is also dose-dependent.
There is some evidence that propofol also causes neuro-apoptosis when used in a critical period of brain development [64]. As with many other sedatives there is currently no certainty regarding the neurodevelopmental consequences.
Conclusions
Propofol is a widely used sedative in adult ICUs, but its use is limited in pediatric patients mainly because of the risk of developing the potentially fatal complication PRIS. Therefore it should not be used for long-term sedation in critically ill children. On the other hand, propofol seems to be a good choice for short-term or postoperative sedation, when the child is likely to be weaned off mechanical ventilation within 24 h. It also is an excellent choice for procedural sedation in children undergoing deep-sedation for medical procedures, including non-intubated patients.
Monitoring Depth of Sedation and Analgesia
Undersedation as well as oversedation can be harmful to the patient and need to be avoided. In order to tailor sedative drugs according to the individual optimal sedation level, the careful assessment of the sedation level in each individual patient at any time of ICU management is extremely important. Different tools are available for assessing the quality and depth of sedation in pediatric ICU patients. In this chapter we will discuss the most commonly applied tools: the COMFORT scale , the COMFORT behavior scale , the State behavioral scale, and the Ramsay scale [48].
The COMFORT scale was developed as a nonintrusive measure for assessing distress in children admitted to pediatric intensive care units. Eight dimensions of behavioral and/or physiologic distress were selected based upon a literature review and survey of PICU nurses [5]. After the introduction of this scale Marx et al. found that adequacy of sedation is measured more consistently by observers using the COMFORT scale than by intensivist global assessment [104].
An important drawback of this scoring system is the fact that it incorporates the physiologic parameters hearth rate and blood pressure which can be influenced by many more factors than distress alone. This has been confirmed by the study of Ista et al. who showed that physiologic variables do not correlate well with the behavioral items of the COMFORT scale . For this reason an abbreviated COMFORT scale restricted to behavior items was developed named the COMFORT behavior scale . It has proven a reliable alternative to the original COMFORT scale [105]. A recent study has shown that the COMFORT behavior scale detects treatment-related changes in pain or distress intensity. This implies that it can effectively guide both analgesic and sedation treatment in critically ill children [4]. The advantage of both the COMFORT and the COMFORT behavior scales is that it is suitable for the assessment of both pain and non-pain related distress in critically ill children. Furthermore they are age independent.
The state behavioral scale was developed to assess sedation and agitation in young (6 weeks to 6 years) pediatric patients on mechanical ventilation [106]. Its main drawback therefore is that it is only suitable for this age group.
The Ramsey scale was originally developed to determine the level of sedation in adult patients admitted to the ICU [107]. It classifies the consciousness into six categories. Although it’s widely used in adults, it has not been validated in children. The same holds true for the Richmond agitation sedation scale (RASS) [108].
In conclusion the COMFORT and COMFORT behavior scale seem to be the most suitable tools for assessing sedation because they have been extensively investigated and assess both pain and non-pain related distress. When using other scoring systems for sedation it is therefore of the utmost importance to assess pain in addition because the scoring systems only assess sedation. In paralyzed children assessing sedation levels is very difficult. In this patient category the PICU modified comfort sedation scale for muscle relaxed patients can be used [109]. For an extensive evaluation of instruments for scoring pain, non-pain related distress and adequacy of analgesia and sedation in ventilated children we refer to the recent review of Dorfman et al. [110].
When assessing comfort of children admitted to an intensive care unit it is very important to consider that pediatric delirium is a frequent occurring phenomenon. Therefore besides assessing the level of sedation and pain it is important to incorporate delirium screening into daily practice. This item is discussed separately in this chapter.
Titrating the depth of sedation thus will ideally prevent both under and oversedation. The question is, however, whether it also will lead to a reduction in duration of mechanical ventilation and/or length of stay. Reports on this subject are contradictory. A recent large trial among children undergoing mechanical ventilation for acute respiratory failure in which the use of a sedation protocol was compared with usual care did not reduce the duration of mechanical ventilation [111]. Another study showed that implementing a sedation protocol for mechanically ventilated children reduced the duration of sedation but only tended to reduce the duration of mechanically ventilation [30]. Reduction of the cumulative dose of benzodiazepines is also shown by another study in which a nurse-driven pediatric analgesia and sedation protocol was studied. In this study also a reduction in the occurrence of withdrawal symptoms was observed [112].
Delirium in the Critically Ill Child
In the last few years there has been a growing awareness of the clinical importance of pediatric delirium (PD) in critically ill children [7, 48, 113]. The prevalence in this patient category (i.e., critically ill children) ranges from 11 to 21 % [114–116].
Delirium is a neuro-cognitive disorder due to a somatic illness or its treatment and is characterized by loss of attention, interests, appetite, emotional irritability, tiredness, and increased need of sleep. According to DSM-5 (American Psychiatric Association, 2013) the four essential features of delirium are:
A disturbance of attention or awareness
This disturbance is accompanied by changes in cognition that cannot be better accounted for by another pre-existing neuro-cognitive disorder (e.g., mental retardation)
The condition develops in a short period of time, hours or days, and often fluctuates during the day, typically worsening in the evening
There are indications from the patients’ history, examination or laboratory results that the disturbance is probably the result of a medical condition or its treatment [117].
This definition officially has a fifth criterion, which has been criticized because of the requirement to exclude cases of reduced level of arousal (“coma”) without specifying how this has to be done [118]. ICD-10 defines delirium as an etiologically nonspecific organic cerebral syndrome characterized by concurrent disturbances of consciousness and attention, perception, thinking, memory, psychomotor behavior, emotion, and the sleep–wake schedule. The duration is variable and the degree of severity ranges from mild to very severe [119].
Delirium has three subtypes: hyperactive (acute agitation, anxiety, hallucinations, delusions), hypoactive (apathy, empty gaze, and formal thought and speech disturbances), and mixed [120].
There are many important reasons for the systematical assessment of PD in the pediatric intensive care unit:
Delirium is “acute brain failure ,” and as the brain also is the director of the autonomic- and endocrine systems the consequences may be severe leading to increased morbidity and mortality.
A hyperactive delirium is accompanied by various risks, such as pulling out IV lines and catheters, auto-extubation, stepping or falling out of bed, etc.
It is stressful for the patient who may experience terrifying hallucinations or delusions (sometimes without amnesia) that may lead to a post-traumatic stress disorder (PTSD) and
It can also be very stressful for the child’s family and clinical staff. Up to 25 % of parents of children who have been in a pediatric ICU may develop PTSD [121].
In general, critically ill children with delirium have a higher resilience (and better prognosis) than adults. The negative neuro-cognitive effects of delirium in adults and the elderly are well known, but we do not know yet whether this is also true for children.
The acute occurrence of a disturbance of cognition, emotions, consciousness, or behavioral disturbances in a critically ill child should raise the suspicion of PD. Nurses and physicians may find it difficult to assess symptoms of delirium, including cognitive changes, especially in pre-verbal, critically ill, and mechanically ventilated children. In these children, other aspects, such as behavioral characteristics and non-verbal interactions between parent and child, should be considered.
Answering the following questions based on one of the delirium screening tools (CAPD) can be helpful in raising the possibility of delirium [122]:
Does the child make eye contact?
Are the child’s actions purposeful?
Is the child aware of its surroundings?
Does the child communicate needs and wants?
Is the child restless?
Is the child inconsolable?
Is the child underactive (moves very little while awake)?
Does the child take a long time to respond to interactions?
The diagnosis of delirium in children older than 5 years with normal development is based on DSM-5 or ICD-10 criteria. Accurately diagnosing pediatric delirium requires using a reliable, valid, and clinically suitable bedside tool that may also serve for screening and to guide treatment such as the Pediatric Anesthesia Emergence Delirium Scale (PAED) , the pediatric Confusion Assessment Method for ICU (PCAM-ICU) , the Cornell Assessment Pediatric Delirium tool (CAPD) , and the Sophia Observation Withdrawal Symptoms-Pediatric Delirium scale (SOS-PD) [115, 122–129].
It must be emphasized that the right diagnosis and treatment of PD cannot be made merely on the basis of observational screening tools. It is mandatory to evaluate all other differential diagnostic explanations. The underlying somatic differential diagnosis of conditions potentially leading to delirium is described in the section refractory agitation in this chapter—see below.
The differential diagnosis of particularly hypoactive delirium , given its high prevalence and frequently disappointing response to treatment deserves particular attention:
Withdrawal of medicationRelated posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
