Following LTR, continuous sedation has been advocated in infants and young children to avoid movement and trauma from the ETT against the fresh graft site and avoid the potentially life-threatening complication of inadvertent tracheal extubation.[5
] Sedative and analgesic agents, although mandatory following surgical procedures and during ongoing endotracheal intubation, may result in adverse respiratory effects including depressed cough reflex, ineffective clearing of secretions, diminished sigh volumes, decreased functional residual capacity as well as hemodynamic effects which may not be well tolerated in the immediate postoperative period or in infants with co-morbid disease processes. Following the prolonged administration of these agents, their abrupt discontinuation may result in a withdrawal syndrome which may complicate the process of tracheal extubation. Despite the beneficial effects of adequate sedation, tolerance and physical dependency to these medications develops in up to 50–60% of patients.[13
] Treatment strategies and protocols are necessary so that the problems of tolerance, physical dependency, and withdrawal do not limit the administration of these agents in this select patient population. The manifestations of withdrawal vary according to the agent used for sedation, manifesting shortly after discontinuation of the drug if the agent has a short half-life (propofol, fentanyl or morphine) or days later if the agent or its metabolites have long half-lives (diazepam).[13
] As tolerance develops related to receptor occupancy, it is theoretically possible to delay its development by using agents with decreased agonism at the receptor or by the use of rotating sedation regimens at specific intervals. One of the other strategies suggested in patients receiving long-term continuous sedation can be replacing synthetic opioids such as fentanyl with non-synthetic opioids such as morphine, as synthetic opioids have increased affinity for the opioid receptor, which may result in tolerance more rapidly and a higher incidence of withdrawal than non-synthetic opioids.[13
] In a study comparing the effects of morphine and fentanyl on the prevalence of withdrawal after extracorporeal membrane oxygenation, Franck et al
. demonstrated that neonates receiving morphine required less supplemental analgesia than did neonates who received fentanyl and had a significantly lower incidence of withdrawal.[15
] This resulted in a more rapid hospital discharge in neonates receiving morphine than those who had received fentanyl. Other potential advantages of morphine are illustrated by additional reports demonstrating not only efficacy, but also beneficial physiologic effects.[16
] Lynn et al
. demonstrated that morphine infusions of 10–30 µg/kg/hour provided effective analgesia without affecting weaning from mechanical ventilation following cardiac surgery in neonates and infants.[16
] Morphine infusions are also effective in blunting the sympathetic stress response and reducing epinephrine adrenaline levels during mechanical ventilation.[17
] In their study of 41 mechanically ventilated babies who were treated with surfactant for hyaline membrane disease, Quinn et al
. demonstrated that morphine-treated neonates had a significant reduction in plasma epinephrine concentrations without adverse hemodynamic effects.[17
To date, there are limited data to demonstrate the efficacy of a rotating sedation regimen in preventing tolerance and thereby limiting the incidence of withdrawal. Wheeler et al
. presented preliminary data on a strategy employing a rotating sedation regimen, derived from the treatment of two patients who required sedation following LTR.[14
] The sedation regimen included a midazolam infusion (0.1–0.15 mg/kg/hour) with as needed doses of morphine on day 1, a fentanyl infusion (2–3 µg/kg/hour) with as needed doses of lorazepam on day 2, and a dexmedetomidine infusion (0.25–0.3 µg/kg/hour) with as needed doses of morphine on day 3. The day 1 regimen was repeated on day 4, the day 2 regimen on day 5, and the day 3 regimen on day 6 so that tracheal extubation occurred during the dexmedetomidine infusion. The authors compared the development of tolerance, physical dependency, withdrawal and hospital discharge in the two patients treated with a rotating sedation regimen with five patients who received the conventional regimen of a continuous infusions of midazolam and intermittent doses of morphine for their entire 5 day postoperative course following LTR. The authors noted no clinical signs or symptoms of withdrawal in the two patients treated with the rotating sedation regimen, whereas all five patients who received continuous infusion of a single agent manifested either mild or moderate withdrawal. The problems of tolerance and withdrawal manifested by the patients who had not received the rotating regimen resulted in the need for a longer hospital stay (mean of 8.6 days versus 6.5 days) and a delay in the time for the resumption of full oral fluid intake.
In addition to issues such as tolerance, withdrawal, and physical dependency which may lead to problems following tracheal extubation, the prolonged effects of sedative and analgesic agents may compromise upper airway control and ventilator function, leading to respiratory failure following tracheal extubation. Patients less than 4 years of age are more vulnerable to prolonged sedation, with a resultant increase in failure rates following tracheal extubation.[2
] Although short-acting agents such as fentanyl and midazolam are frequently chosen for sedation during mechanical ventilation, significant changes in their elimination half-life may occur with prolonged administration (context-sensitive half-life) so that a prolonged effect is noted following their discontinuation.
One of the means of limiting residual sedation includes titration of infusion rates to the desired level of sedation by following pediatric pain scores.[18
] The currently used PICU sedation scores evaluate either physiologic variables such as heart rate and blood pressure, an objective assessment of the patient’s depth of sedation, or a combination of the two. One commonly used scale, the COMFORT score, combines the scoring of a patient’s response or movement in addition to various physiologic parameters.[19
] It relies on the measurement of alertness, respiration, blood pressure, muscle tone, agitation, movement, heart rate, and facial tension. This scoring system has been validated in the pediatric-aged patient and may have utility in the assessment of sedation during mechanical ventilation.[19
] However, scales that use physiologic parameters can be misleading in an ICU setting where alterations in vital signs can occur unrelated to the level of sedation. Furthermore, patients with cardiovascular dysfunction requiring vasoactive medications may not manifest increases in heart rate and blood pressure even in the presence of severe agitation or pain. Because of these concerns, Ista et al
. have recently proposed a modified original COMFORT score, known as the COMFORT-B score which eliminates the use of physiologic variables and provides new cutoff points for the diagnosis of oversedation or undersedation.[21
] Other scoring systems, such as the Sedation-Agitation Scale (SAS), also eliminate the use of physiologic parameters. The SAS visually assesses the level of the patient’s comfort and grades it from 1 (unarousable) to 7 (dangerous agitation such as pulling at the ETT).[22
] The Ramsay Scale, a sedation scale used commonly in the adult ICU population, not only assigns a value based on the observation of the patient but also uses a tactile stimulus (a glabellar tap) to distinguish between the deeper levels of sedation.[23
] Scoring for the Ramsey score varies from 1 (awake, anxious and agitated) to 6 (no response to a glabellar tap). The Hartwig score similarly uses a visual assessment of the patient, but as with the Ramsay scale, includes a response to a noxious stimulus (in this case, tracheal suctioning), thereby eliminating its use in non-intubated patients.[24
] Scales such as the Ramsay and the Hartwig that assess the response to a tactile stimulus require disturbing the patient to differentiate between the deeper levels of sedation. Additionally, scales that evaluate a patient’s response to a stimulus or observe their behavior are not valid during the administration of NMBAs which prevent movement.
As none of these sedation scales meets all of the needs of the PICU provider, there remains an interest in the use of monitoring technology which may be able to assess the depth of sedation through the analysis of the electroencephalogram (EEG). The Bispectral Index (BIS monitor) (Aspect Medical, Newton, MA, USA) uses a programmed algorithm to evaluate the processed EEG pattern and provide a numeric value ranging from 0 (isoelectric) to 100 (awake with eyes open). Its predominant clinical use has been intraoperatively to monitor the effects of general anesthetic and sedative agents and provide a measure of the depth of anesthesia. Although still somewhat controversial, it has been suggested that maintenance of a BIS value to less than 60–70 correlates with a low probability of intraoperative awareness.[25
] Caveats regarding the use of BIS are that the algorithm was developed during the use of general anesthetic agents such as propofol, barbiturates, benzodiazepines or the inhalational anesthetic agents which act through the γ-amino butyric acid (GABA) system. Additionally, as there are differences between the EEG of the adult and children less than 6-8 years of age, these devices may have limited utility in younger patients. Although the results have been somewhat mixed, the majority of reports have demonstrated a clinically acceptable correlation between the BIS monitor and commonly used ICU sedation scores including the Ramsay or the COMFORT score.[27
] Although clinical pain scores are generally quite useful to allow for the effective titration of the sedation regimen, when pain scores are not feasible such as during the use of NMBAs, processed EEG monitoring may be helpful in the evaluation of the depth of sedation.
Additional means of limiting the residual effects of sedative and analgesic agents during prolonged infusions in the PICU include the use of intermittent dosing rather than a continuous infusion or the use of “drug holidays” as have been popular in the care of adult ICU patients. In a landmark study by Kress et al
., daily interruption of sedative drug infusions decreased the duration of mechanical ventilation and the length of stay in the intensive care medicine in their cohort of 128 patients.[32
] The median duration of ventilation and median length of stay in ICU was significantly reduced in the interventional group receiving daily interruptions in sedative infusions (4.9 versus 7.3 days and 6.4 versus 9.9 days, respectively).[32
] In addition to trying to avoid or limit physical tolerance and withdrawal, successful tracheal extubation can be facilitated by avoiding residual effects of sedative agents which may impair upper airway control and respiratory function. To accomplish this, switching to agents which do not exhibit changes in their duration or effect following prolonged continuous infusions may also be beneficial. Therefore, it may be beneficial to switch to propofol or remifentanil for 8–12 hours prior to an anticipated attempt at tracheal extubation. As these agents do not exhibit a significant context sensitive half-life, especially remifentanil, their effects should dissipate rapidly upon discontinuation of the infusion. Given the potential for the development of the propofol infusion syndrome with the prolonged continuous infusion of propofol, this agent has been eliminated from the armamentarium of many PICUs.[33
] However, the overwhelming majority of the severe cases of propofol infusion syndrome occurred with infusions of greater than 24–48 hours and, therefore, short-term infusions of 8–12 hours to allow for the clearance of the residual effects of other agents may be acceptable. However, it may be prudent to periodically monitor acid–base status and lactate levels even during the short-term administration of propofol and discontinue its administration immediately should a lactic acidosis develop as this may herald the onset of the other manifestations of the propofol infusion syndrome. Alternatively, the short acting synthetic opioid, remifentanil, may also be used to provide a deep level of sedation with a rapid offset once the infusion is discontinued. Remifentanil is metabolized by plasma esterases and demonstrates stable and similar pharmacodynamics across all age ranges.[18
Its half-life of 8–10 minutes is constant even following prolonged administration and anecdotal experience has demonstrated that it may provide a deep level of sedation without residual effects when the infusion is discontinued.[35
] The disadvantages of remifentanil which mandate its restriction to short-term (less than 24 hours) use are its cost and the rapid development of tolerance with the resultant need to rapidly increase the dose to maintain the same level of sedation.
Adjunctive agents for the perioperative care of these patients, such as dexmedetomidine, diphenhydramine, clonidine, phenothiazines, non-steroidal anti-inflammatory agents, acetaminophen, and chloral hydrate, have also been suggested by many authors to limit the need for opioids and benzodiazepines in an attempt to prevent or limit the development of tolerance, physical dependency, and withdrawal.[1
] As these agents have limited effects on respiratory function, they may provide significant benefit particularly when used in rotation with opioids and benzodiazepines. Dexmedetomidine is an α2
-adrenergic receptor agonist that possesses sedative, analgesic, and anxiolytic properties with no limited effects on respiratory function when administered within clinical dosing guidelines.[37
] The short half-life of dexmedetomidine (~2 hours) allows easy titration by continuous infusion, quicker recovery, and fewer prolonged sedation related adverse effects. Given the concerns of respiratory depression, hemodynamic instability, and metabolic acidosis associated with the administration of propofol, dexmedetomidine may be a suitable alternative to allow for rapid tracheal extubation following prolonged sedation with opioids and benzodiazepines.
Although the majority of studies demonstrate a favorable pattern of hemodynamic stability of dexmedetomidine in pediatric patients, dexmedetomidine has the potential to produce dose-dependent decreases in blood pressure and heart rate.[39
] Rarely, dexmedetomidine has also been reported to cause life-threatening complications including sinus arrhythmias, left ventricular dysfunction, refractory cardiogenic shock, and cardiac arrest.[37
] Additionally, although approved for sedation during mechanical ventilation of adults, dexmedetomidine has not been approved by the Food and Drug Administration (FDA) for use in infants and children.[43
Dexmedetomidine dosing regimens have been extrapolated from the adult literature, with modifications based on clinical experience in pediatric-aged patients. Current recommendations include a bolus dose of 0.5–1 µg/kg administered over 10 minutes, followed by a continuous infusion of 0.2–1.5 µg/kg/hour.[18
] Dexmedetomidine has also been used during the extubation process to provide sedation and anxiolysis with limited effects on ventilatory function. It does not appear to significantly depress respiratory drive, thus interference with weaning from mechanical ventilation is less likely. In fact, it has been used both as a bridge to extubation and to expedite the process of weaning from mechanical ventilation.[44
] In a study by Arpino et al
., dexmedetomidine was initiated in a group of mechanically ventilated patients who failed previous attempts at weaning and tracheal extubation secondary to agitation.[44
] With the administration of dexmedetomidine, 65% of the patients were able to undergo successful tracheal extubation. Dexmedetomidine was associated with a reduction in concomitant sedative and analgesic use with minimal adverse effects.
Non-steroidal anti-inflammatory agents and acetaminophen may be useful to provide adjunctive analgesia following these surgical procedures and thereby limit the use of opioid agents and their associated adverse effects. Although the non-steroidal anti-inflammatory agent, ketorolac, has been available for intravenous administration for years, recent additions to our practice include the availability of intravenous ibuprofen and the upcoming release of an intravenous acetaminophen preparation. Several studies in the pediatric population have demonstrated that these agents can effectively reduce opioid requirements by up to 20–30% following major surgical procedures. Additionally, the use of non-pharmacologic measures such as reduced stimulation, comforting, and regulation of day–night cycle are the other suggested methods to reduce the amount of sedative medications.[1