Analgesia, sedation, and anesthesia are a continuum. Diagnostic and/or therapeutic procedures in newborns often require analgesia, sedation, and/or anesthesia. Newborns, in general, and, particularly, those with heart disease, have an increased risk of serious adverse events, including mortality under anesthesia. In this section, we discuss the assessment and management of pain and discomfort during interventions, review the doses and side effects of commonly used medications, and provide recommendations for their use in newborns with heart disease. For procedures requiring deeper levels of sedation and anesthesia, airway and hemodynamic support might be necessary. Although associations of long-term deleterious neurocognitive effects of anesthetic agents have received considerable attention in both scientific and lay press, causality is not established. Nonetheless, an early multimodal, multidisciplinary approach is beneficial for safe management before, during, and after interventional procedures and surgery to avoid problems of tolerance and delirium, which can contribute to long-term cognitive dysfunction.
In the past several years, there are data that now allow a better appreciation of neonatal pain responses and therapeutic options required to sustain an appropriate degree of safe and effective pain control. The identifiable and specific neonatal reactions to therapeutic cardiac interventions necessitate a deep understanding of the potential medications and balancing their desired effects with their potential negative side effects. This review of the data and literature provides a foundation for a successful analgesic and pain control care plan in neonatal cardiac patients.
This article is part of a larger series of articles simultaneously published as a Supplement in Pediatrics by the Neonatal Cardiac Care Collaborative. Please refer to the Executive Committee introductory paper for a discussion on Class of Recommendations and Level of Evidence (LOE), writing committee organization, and document review and approval.
Preoperative Neonatal Sedation and Analgesia
For neonates with congenital heart disease, the use of sedation and analgesia in the preoperative period is mainly reserved for the management of pain during noxious procedures. In addition, it may be necessary for the provision of comfort with safe maintenance of the airway for intubated neonates and for the reduction of metabolic demand for those at risk for hemodynamic collapse, such as neonates with single ventricle physiology at risk for overcirculation (ie, systemic underperfusion due to excessive pulmonary blood flow) and those with pulmonary hypertension. Sedatives and analgesics are thought to decrease tissue metabolic demand through a reduction in the neuro-humoral response to stress, decreased spontaneous muscular activity, a reduction in work of breathing, and/or decreased body temperature.1
Assessment of Analgesia and Sedation
Although it is firmly believed that neonates can experience pain and require appropriate pain-relieving treatment, it is often difficult to assess them for adequate pain relief or sedation because of their preverbal state. More than 40 pain assessment instruments for neonates have been developed, but not all share the same degree of psychometric rigor, and many are designed for acute procedural pain only.2 The European Society of Pediatric and Neonatal Intensive Care (ESPNIC) formulated a position statement guiding the assessment and reassessment of the effectiveness of pain and sedation interventions useful for both pre- and postoperative management. Table 1 summarizes the most recommended pain and sedation assessment tools per the ESPNIC guideline.3 The Neonatal Pain, Agitation, and Sedation Scale attempts to assess both acute and prolonged pain in addition to distress with prolonged pain defined as pain primarily caused by a disease process or noxious stimuli such as the presence of an endotracheal tube or chest tube.3,4 The COMFORTneo scale is an adaptation of the COMFORT behavior scale that was generally validated in a large cohort of neonates and intended to better represent the characteristics of preterm neonates.5 Unlike other assessment tools focused on procedural pain, the COMFORTneo scale assesses prolonged pain and also attempts to identify oversedation.5 The State Behavioral Scale (SBS) is a tool intended to describe sedation/agitation levels in intubated children on a ventilator.6 Although the SBS has not been validated in neonates, it is still sometimes used in this population, especially in units in which the age span of patients is broader than that of a traditional NICU.
| PIPP | PIPP-revised | N-PASS | COMFORTneo | SBS | |
|---|---|---|---|---|---|
| Age | 28–40 wk | 28–40 wk | 23–40 wk | 24–42 wk | 6 wk–6 y |
| Type of assessment | Procedural & postoperative pain | Procedural pain | Procedural & prolonged pain | Prolonged pain | Sedation |
| Variables assessed | Heart rate | Heart rate | Heart rate | Alertness | Respiratory drive |
| Oxygen saturation | Oxygen saturation | Respiratory rate | Calmness/agitation | Coughing | |
| Brow bulge | Brow bulge | Blood pressure | Respiratory response or crying | Best response to stimuli | |
| Eye squeeze | Eye squeeze | Oxygen saturation with stimuli | Body movement | Attentiveness to care provider | |
| Nasolabial furrow | Nasolabial furrow | Facial expressions | Muscle tone | Tolerance to care | |
| Behavioral state | Behavioral state | Behavioral state | Facial tension | Consolability | |
| Extremities/tone | Movement after consoled | ||||
| Score range | 0–21 | 0–21 | 4-point scale | 6–30 | 6-point scale |
| 0–6 None to mild | 0–6 None to mild | −2 to +2 with | 6–13 None to mild | −3 to +2 with | |
| 7–12 Moderate | 7–12 Moderate | 0 = awake and calm | 14–21 Moderate | 0 = awake and calm | |
| >12 Severe | >12 Severe | >22 Severe | |||
| Adjustment for gestational age | Yes | Yes | Yes | No | No |
| PIPP | PIPP-revised | N-PASS | COMFORTneo | SBS | |
|---|---|---|---|---|---|
| Age | 28–40 wk | 28–40 wk | 23–40 wk | 24–42 wk | 6 wk–6 y |
| Type of assessment | Procedural & postoperative pain | Procedural pain | Procedural & prolonged pain | Prolonged pain | Sedation |
| Variables assessed | Heart rate | Heart rate | Heart rate | Alertness | Respiratory drive |
| Oxygen saturation | Oxygen saturation | Respiratory rate | Calmness/agitation | Coughing | |
| Brow bulge | Brow bulge | Blood pressure | Respiratory response or crying | Best response to stimuli | |
| Eye squeeze | Eye squeeze | Oxygen saturation with stimuli | Body movement | Attentiveness to care provider | |
| Nasolabial furrow | Nasolabial furrow | Facial expressions | Muscle tone | Tolerance to care | |
| Behavioral state | Behavioral state | Behavioral state | Facial tension | Consolability | |
| Extremities/tone | Movement after consoled | ||||
| Score range | 0–21 | 0–21 | 4-point scale | 6–30 | 6-point scale |
| 0–6 None to mild | 0–6 None to mild | −2 to +2 with | 6–13 None to mild | −3 to +2 with | |
| 7–12 Moderate | 7–12 Moderate | 0 = awake and calm | 14–21 Moderate | 0 = awake and calm | |
| >12 Severe | >12 Severe | >22 Severe | |||
| Adjustment for gestational age | Yes | Yes | Yes | No | No |
PIPP, Premature Infant Pain Profile; N-PASS, Neonatal Pain, Agitation, and Sedation Scale
This was table modified from The European Society of Pediatric and Neonatal Intensive Care Guidelines.3
Many institutions now use a local iteration of a goal-directed, nurse-implemented sedation protocol with the aim of optimizing sedation and analgesia while avoiding unnecessary doses. The impact of implementing these protocols is mixed, with some studies revealing no difference in time to extubation, ICU or hospital length of stay, or incidence of withdrawal, whereas others reveal a significant decrease in sedative and/or analgesic exposure with a concomitant decrease in the incidence of withdrawal without an increase in negative side effects such as unintentional extubation, inadequate pain control, or agitation.7–9 Nonpharmacologic strategies, such as minimizing noxious environmental stimuli (temperature, noise), swaddling, or nonnutritive sucking, are also important adjuncts to the pharmacologic management of sedation and analgesia.3,10 The commonly used medications are briefly described below.
Opioids
Although primarily indicated for analgesia, opioids are also often first-line agents for sedation. The analgesic effect of opioids occurs primarily through binding to μ opioid receptors.11 Table 2 summarizes commonly used opioids. Opioid pharmacokinetics in neonates are impacted by a low plasma protein concentration and a higher body water composition, which affects drug distribution. An immature metabolic process, such as low hepatic enzyme activity, affects clearance, and a reduced renal excretion secondary to immature glomerular filtration, tubular secretion, and reabsorption prolongs the half-life.12 Continuous infusions have traditionally been used to avoid large variations in plasma concentration, but several studies suggest that intermittent dosing is an effective strategy in providing adequate analgesia and sedation while reducing the cumulative opioid dose.13,14 Although oxycodone is often used in older children and adults, high individual variability in metabolism and longer clearance times, especially in premature infants, limits its use in the neonatal population.15
Commonly Used Opioids
| Drug | Pharmacokinetics | Usual Starting Dose | MME | Comments |
|---|---|---|---|---|
| Morphine | Half-life: 2–3 h | 0.05–0.1 mg/kg/dose IV | 1 | Avoid in chronic kidney disease |
| Infusion (IV): 10–50 µg/kg/hr | ||||
| 0.3 mg/kg/dose PO | ||||
| Fentanyl | Half-life: 0.3–0.5 h | 0.5–2 µg/kg/dose IV | 0.13–0.18 | Avoid rapid push because of risk for chest wall rigidity |
| Infusion (IV): 0.2–2 µg/kg/hr | ||||
| Hydromorphone | Half-life: 2–4 h | 0.015 mg/kg/dose IV | 4 | Reduce dose in patients with renal disease |
| 0.03 mg/kg/dose PO |
| Drug | Pharmacokinetics | Usual Starting Dose | MME | Comments |
|---|---|---|---|---|
| Morphine | Half-life: 2–3 h | 0.05–0.1 mg/kg/dose IV | 1 | Avoid in chronic kidney disease |
| Infusion (IV): 10–50 µg/kg/hr | ||||
| 0.3 mg/kg/dose PO | ||||
| Fentanyl | Half-life: 0.3–0.5 h | 0.5–2 µg/kg/dose IV | 0.13–0.18 | Avoid rapid push because of risk for chest wall rigidity |
| Infusion (IV): 0.2–2 µg/kg/hr | ||||
| Hydromorphone | Half-life: 2–4 h | 0.015 mg/kg/dose IV | 4 | Reduce dose in patients with renal disease |
| 0.03 mg/kg/dose PO |
MME, morphine mg equivalent; IV, intravenous; PO, by mouth.
This table was modified from Parikh et al.12
For patients with hemodynamic lability or those at risk for pulmonary hypertension, fentanyl may be a better choice given its ability to blunt the sympathetic stress response with less histamine release than morphine.16 However, caution must be used in intermittent dosing of fentanyl because of the potential for developing chest wall rigidity. Should this adverse event occur, the effect can be reversed with naloxone or interrupted by a neuromuscular blockade. For most neonates, morphine is an acceptable alternative, especially in those with normal cardiac function. However, veno-dilation by morphine may lead to hypotension, especially in a hypovolemic patient.16 Opioids do not provide amnesia. Side effects include respiratory depression, nausea, vomiting, constipation, and urinary retention. Prolonged use of opioids is associated with increased tolerance, requiring escalation of the dose to achieve the same analgesic or sedative effect. This has particularly been well-described with prolonged use of a fentanyl infusion.17 Neonates are also susceptible to opioid withdrawal, which can be attenuated by a slow taper and/or the introduction of a longer-acting agent, such as methadone. Although there is concern that the exposure of the developing neonatal brain to opioid infusions beyond 5 to 7 days may impair short-term memory, lead to poorer motor development, and result in more social problems compared with placebo-treated children, other studies have refuted these findings.18,19
Benzodiazepines
Until the recent scrutiny of benzodiazepines in their contribution to delirium, these agents were the most commonly used sedatives in many intensive care units.16 Through their impact on the inhibitory neurotransmitter γ-aminobutyric acid, particularly in the limbic system, benzodiazepines induce anterograde amnesia. Benzodiazepines do not provide analgesia and are often combined with opioids for perceived pain.20 Neonates have fewer γ-aminobutyric acid-A receptors, increasing the risk of neuro-excitability and clonic activity resembling seizures with prolonged benzodiazepine exposure.10 In premature infants (24–32 weeks’ gestation), benzodiazepines can also be associated with worse short-term adverse effects, including severe intraventricular hemorrhage and periventricular leukomalacia.21
The 3 most common benzodiazepines used include lorazepam, midazolam, and diazepam. Diazepam’s long half-life often limits its use in the neonatal population (Table 3). Given midazolam’s short half-life, it is usually administered as a continuous infusion. Other medications commonly used in the ICU, such as heparin, may decrease the protein binding of midazolam, increasing the free fraction of the drug. Hepatic and renal dysfunction may also contribute up to 2.5 to 3 times the increased free fraction of midazolam.22,23 Unfortunately, exposure to benzodiazepines beyond 5 to 7 days is associated with both tolerance and withdrawal symptoms that can be attenuated by slowly tapering the dose or transitioning to an oral agent with a longer half-life (eg, midazolam to lorazepam).24
| Drug | Pharmacokinetics | Usual Starting Dose | Comments |
|---|---|---|---|
| Lorazepam | Half-life: 4–12 h | Intermittent IV: 0.05–0.1 mg/kg every 4–8 h | Risk for propylene glycol accumulation if used as prolonged infusion and metabolic acidosis |
| Metabolism via glucuronyl transferase | Continuous infusion (IV): 0.025 mg/kg/hr | Not impacted by other medications that alter the P450 system | |
| No active metabolites | |||
| Diazepam | Half-life: 12–12 h | IV: 0.05–0.1 mg/kg (max: 0.25 mg/kg) | Hypotension/apnea associated with rapid infusion |
| Hepatic metabolism | PO: 0.2–0.3 mg/kg | Not recommended for continuous infusion | |
| Active metabolites | |||
| Midazolam | Half-life: 2–4 h | PO: 0.5–0.7 mg/kg | Prolonged effect with hepatic or renal dysfunction |
| Hepatic metabolism | Rectal: 1.0 mg/kg | ||
| Renal excretion | Continuous infusion (IV): 0.05–0.2 mg/kg/hr | ||
| Active metabolite may accumulate with prolonged infusion |
| Drug | Pharmacokinetics | Usual Starting Dose | Comments |
|---|---|---|---|
| Lorazepam | Half-life: 4–12 h | Intermittent IV: 0.05–0.1 mg/kg every 4–8 h | Risk for propylene glycol accumulation if used as prolonged infusion and metabolic acidosis |
| Metabolism via glucuronyl transferase | Continuous infusion (IV): 0.025 mg/kg/hr | Not impacted by other medications that alter the P450 system | |
| No active metabolites | |||
| Diazepam | Half-life: 12–12 h | IV: 0.05–0.1 mg/kg (max: 0.25 mg/kg) | Hypotension/apnea associated with rapid infusion |
| Hepatic metabolism | PO: 0.2–0.3 mg/kg | Not recommended for continuous infusion | |
| Active metabolites | |||
| Midazolam | Half-life: 2–4 h | PO: 0.5–0.7 mg/kg | Prolonged effect with hepatic or renal dysfunction |
| Hepatic metabolism | Rectal: 1.0 mg/kg | ||
| Renal excretion | Continuous infusion (IV): 0.05–0.2 mg/kg/hr | ||
| Active metabolite may accumulate with prolonged infusion |
IV, intravenous; PO, by mouth.
Increasing attention has been given to the association of prolonged benzodiazepine use with delirium. The authors of a retrospective study reported a temporal, causal, and dose-dependent relationship between benzodiazepine use and the development of delirium.25 Given this heightened concern, several institutions have shifted their sedation practices to a benzodiazepine-sparing approach, with the ability to successfully reduce the cumulative dose of benzodiazepine while still safely achieving target sedation scores without negative side effects.8,26,27
Despite animal studies suggesting a detrimental impact of sedation and analgesic agents on long-term neurodevelopmental outcomes, current studies in infants with congenital heart disease do not reveal an association between dose and duration of sedation/analgesic drugs and adverse neurodevelopmental outcomes at 18 to 24 months of age, and only a small statistically significant association between benzodiazepine cumulative dose and lower visual motor integration score at kindergarten age.28,29
α-2 agonists
Dexmedetomidine, a pure α-2-adrenergic agonist provides both sedative and analgesic effects as a solo agent and has both benzodiazepine and opioid-sparing properties as an adjunct agent.30 A typical starting point for infusion is 0.05 µg/kg/hr (range 0.05– 0.2 µg/kg/hr), although several authors also suggest a loading dose of 0.05 µg/kg.14,30 In a study assessing the safety and efficacy of dexmedetomidine in neonates, heart rate and blood pressure decreased during prolonged infusion.30 Caution should be used in patients with relative bradycardia before the initiation of dexmedetomidine or those with second or third-degree heart block.30 Dexmedetomidine is highly protein-bound and a low plasma albumin level results in a greater level of free drug. It is primarily metabolized by the liver and excreted by the kidneys and should be used with caution if liver enzymes are elevated. Preterm infants have a larger volume of distribution with a significantly increased half-life compared with older neonates.30 Unlike opioids and benzodiazepines, dexmedetomidine is not associated with respiratory depression.31
Although clonidine is another α-2-agonist, it is not as selective as dexmedetomidine (α-2/α-1 selectivity ratio of 1620:1 vs 220:1).32 Clonidine has not been extensively studied in the neonatal population but has been shown to decrease the requirement for other sedative agents and attenuate withdrawal symptoms during weaning from benzodiazepine or opioid infusions in older pediatric patients.32 Given its long half-life (12–33 hours, likely longer in neonates) and its propensity for causing hypotension and bradycardia, its use in a preoperative neonate is limited.32 Although there are yet to be long-term studies exploring the impact of α-2 agonists on neurodevelopmental outcomes, 1 study summarizes both in vivo and in vitro neuroprotective mechanisms of dexmedetomidine.33
Other Agents
Ketamine
Ketamine, an antagonist of the N-methyl-D-aspartate receptor, serves as a dissociative anesthetic agent with analgesic, amnestic, and sedative properties. Despite its use in older children, particularly for conscious sedation, few studies have been performed in neonates to ensure an adequate safety profile and, therefore, it is only recommended for highly invasive procedures.10,34 Although some sedation protocols have traditionally included prophylactic benzodiazepine doses to attenuate the dissociative effect of ketamine, data supporting this practice is lacking and should not be used routinely.35
Propofol
Although propofol is an attractive sedative agent given its rapid onset and offset and lack of active metabolites, it is primarily an anesthetic, and it should be used with caution in neonates in the ICU. Clearance of propofol is inversely related to neonatal and postmenstrual age, with significant interindividual variability in pharmacokinetics, increasing the risk for hemodynamic lability and propofol infusion syndrome, characterized by refractory metabolic acidosis, bradycardia, rhabdomyolysis, and cardiovascular collapse.36,37 In the cardiac population, in particular, propofol’s risks are further exacerbated by its negative inotropic effect, significant changes in preload, and potential for increased central vagal tone with subsequent bradycardia.37
Summary: Preoperative Neonatal Sedation and Analgesia
In the preoperative neonate with congenital heart disease, pharmacologic management of sedation and analgesia is an important tool for the patient undergoing noxious stimuli, requiring mechanical ventilation, or in those with high-risk physiology and hemodynamic lability requiring intentionally decreased metabolic demand. Although there is concern for iatrogenic withdrawal, delirium, and potential negative impact on long-term neurodevelopmental outcomes, it is important for the practitioner to carefully consider the risks and benefits of untreated pain or inadequate sedation with these long-term consequences.
Anesthesia for Neonates With Congenital Heart Disease
Anesthesia Management for Cardiac Surgery
The neonate undergoing cardiac surgery with cardiopulmonary bypass (CPB) is an extreme example of a high-risk anesthesia procedure with greater anesthetic morbidity and mortality than other pediatric anesthetic populations. Neonatal cardiac surgery with CPB involves major physiologic changes, which may include deep hypothermia to 18 °C, possible circulatory arrest, and intravascular hemodilution by >50%.
Most neonates will survive after congenital cardiac surgery, but the risk of adverse neurologic outcomes is high.38 The anesthesia management is specific to the patient’s condition and the surgical procedure. Coronary and cerebral perfusion in unrepaired patients with hypoplastic left heart syndrome or truncus arteriosus is critically dependent on balancing systemic and pulmonary vascular resistance. Strategies to increase pulmonary vascular resistance (hypercarbia, acidosis, avoidance of hyperoxia) may be insufficient to manage circulatory instability after anesthesia, and urgent surgical intervention may be required. Preoperative planning is crucial, and many centers schedule multidisciplinary surgical discussion meetings to review each patient’s condition before surgery. General anesthesia is often required for imaging procedures to inform these discussions.
Typically, no premedication is required. Induction of anesthesia can be via inhalation or intravenous methods. Sevoflurane and propofol cause greater myocardial depression than fentanyl, etomidate, or ketamine, although all drugs can cause significant hypotension.39 Patients are intubated with a nondepolarizing muscle relaxant, and maintenance of anesthesia may be a combination of opioid and inhalation anesthesia. α agonists, such as dexmedetomidine, and benzodiazepines, such as midazolam, are possible adjuvants.
Standard anesthesia monitors include 3 or 5 lead electrocardiograms, pulse oximetry, capnography, 2 temperature probes, and an appropriate-sized blood pressure cuff. Monitoring can be difficult because of peripheral vasoconstriction, which reduces the sensitivity of pulse oximetry. External light sources can affect pulse oximetry saturation and near infrared spectroscopy (NIRS) cerebral monitoring. Endotracheal tube and circuit leaks, small tidal volumes, and surgical manipulation in the thoracic cavity can make capnography values difficult to interpret. Frequent correlation with blood plasma carbon dioxide levels is warranted. Most cases will require an intraarterial catheter and an intracardiac pressure line, which may be a central line (femoral, internal jugular, or subclavian vein), peripherally inserted central catheter, or direct transthoracic lines placed at the time of surgery. Cannulation sites for intraarterial access most commonly include the umbilical or radial artery, with alternatives including the ulnar, femoral, or axillary arteries. Confirming the patient’s anatomy, such as subclavian artery aberrancy, or previous surgical manipulation, such as a cutdown, Blalock-Thomas-Taussig shunt, or subclavian flap, is essential before arterial line placement to ensure accurate pressure measurement.40 Ischemic limb damage is rare but a reduction in vessel lumen diameter and/or thrombosis in veins and arteries is common.41 Small cannulas (24G) or peripheral arterial catheters in distal limbs (such as posterior tibial artery) may function poorly after CPB and may not reflect central arterial pressure. Some centers will avoid central venous access in the neck vessels in single ventricle patients to reduce the risk of lumen compromise, which can adversely affect subsequent palliative surgical strategies. Ultrasound guidance can assist in minimizing the time and local trauma in the placement of arterial and central venous lines.41,42
Additional monitoring specific to neonatal cardiac surgery includes neurologic monitoring, such as cerebral oximetry. Some centers also monitor transcranial Doppler and or processed EEG-based technologies with automated interpretation to correlate electrical brain wave activity with depth of anesthesia.43 Both rectal and nasopharyngeal temperature monitoring are used because they reflect core and brain temperature, respectively.
Anesthesia monitors provide important information that can alter surgical decision making. Pulse oximetry can be used as a guide to determine the tightness of pulmonary artery bands or the adequacy of systemic to pulmonary shunts. Hypothermia reduces cerebral blood and spinal cord oxygen consumption, minimizing neurologic injury; hence CPB flow can be reduced. Prolonged lower cerebral oximetry values correlate with poorer neurologic outcomes and perioperative mortality.44 NIRS values have helped identify issues with aortic cannulation placement.44 Intraoperative strategies to minimize neurologic injury include maintaining a hematocrit on bypass ≥25%, avoiding hypoglycemia, limiting deep hypothermic circulatory arrest to 30 minutes, or considering alternatives such as anterograde cerebral perfusion, using NIRS to treat rSO2 ≤50%, and maintaining mean arterial blood pressure on bypass at 40 to 45 mmHg.45
Echocardiography is now routinely performed after open heart surgery involving CPB. A transesophageal probe can be inserted into children as small as 2.8 kg, but the probe can hinder ventilation. Ventilation difficulties or children <2.8 kg may warrant epicardial echocardiography.
Anesthesia for Cardiac Patients Outside the OR
Anesthesia in the cardiac catheterization laboratory (CCL) may be for diagnostic or interventional purposes. Registry data confirms that the CCL is a high-risk environment with neonatal age an independent risk factor for adverse outcomes.46,47 This registry identified a 2.1% all-cause mortality in all patients who had a catheterization procedure during their admission (12% for newborns).46,47
The CCL is a challenging environment to care for ill neonates. Neonates will require intubation. Typically, there are intermittent short periods of intense surgical stimulus on a background of minimal stimulus. Maintaining cardiac output during these periods can be challenging. The laboratory is air-conditioned to cooler temperatures to prevent equipment from overheating. The radiograph arms obscure access to the patient and monitors. Procedures can be several hours and involve ionizing radiation which presents increased risks to neonates.48 Combining MRI with catheterization potentially reduces the risk of radiation at the expense of increased anesthesia exposure. Balancing these risks is challenging because the “upper dose limit” for radiation and anesthesia in the neonate is unknown.
Many cardiac patients undergo cardiac and/or brain MRIs or computed tomography scans for surgical planning or to assess their neurologic state before and after surgery. Neonates under anesthesia will require airway intervention, usually an endotracheal tube, but a laryngeal mask airway may be used for a transient breath hold during a computed tomography scan. Limited patient access and reduced monitoring are characteristic of anesthesia in the MRI suite in which electrocardiogram monitoring can be difficult to interpret because of interference. The MRI is a source of heat, but the MRI room is often cool, hence temperature monitoring is essential. An alternative technique (“feed and sleep”) requires no anesthesia.49 The baby is fasted for 4 hours and then fed and swaddled, which limits movement. Heart and brain imaging can be acquired in 2 stages by using this technique.
In summary, neonates with heart disease may require several trips out of the ICU for diagnostic imaging and/or therapeutic procedures. Each of these requires sedation, anesthesia with or without intubation, and possible use of neuromuscular blocking agents. In addition to the usual risks of transport and common side effects of potent sedating and anesthetic agents, their ability to alter the physiology of the heart disease must be considered. A multidisciplinary approach is required for safety and to minimize adverse outcomes.
Recommendations:
Intraoperative strategies recommended to minimize neurologic injury while on cardiopulmonary bypass include: maintaining a hematocrit on bypass ≥25%, avoiding hypoglycemia, limiting deep hypothermic circulatory arrest to 30 minutes, or considering alternatives such as anterograde cerebral perfusion, using NIRS to treat cerebral somatic regional oxygen saturation ≤50%, and maintaining mean arterial blood pressure on bypass at 40 to 45 mmHg (Class I, LOE B-NR)
Avoidance of arterial access on the same limb as a previous cutdown, subclavian flap for coarctation repair, or a Blalock-Thomas-Taussig shunt is recommended (Class I, LOE B-NR)
Using the feed and sleep technique for diagnostic imaging in newborns is recommended (Class I, LOE B-NR)
Postprocedural Sedation and Analgesia for the Cardiac Newborn
In the early postoperative period, newborns with congenital or acquired heart disease require analgesia and/or sedation to preemptively treat postoperative pain, maximize comfort, reduce the stress response, avoid excessive oxygen consumption, and maintain invasive monitoring devices. The depth of sedation ranges from minimal (anxiolysis), with the maintenance of a natural airway, to deep sedation with depressed consciousness, intubation, and mechanical ventilation. Adequate analgesia/sedation with reduced stress offers an opportunity for improved outcomes.50 Careful titration is required to achieve a balance between undersedation and adverse hemodynamic effects, poor healing and untoward events (self-extubation), and oversedation with hemodynamic instability, delirium, and long-term habituation.51,52 Prolonged opioid/benzodiazepine therapy, defined as >7 days, often leads to tolerance and is associated with withdrawal when weaned or discontinued. Withdrawal can be treated by using a variety of therapeutic approaches, but it may be more desirable to block the mechanisms including high cumulative dose exposure.24
A growing awareness of the long-term deleterious effects on neurodevelopment coupled with the 2016 Food and Drug Administration Drug and Safety Communication and the comprehensive Mayo Anesthesia Safety in Kids trial reported in 2018 regarding anesthetic and sedation medications, including benzodiazepines, has added further emphasis on the need for careful consideration of the use of sedation drugs for the postoperative newborn.53–55
Analgesia and sedation requirements are variable, and no ideal medication or regimen fits every patient, but optimizing practice through validated objective scoring and pathways provides titratable control and improves communication and satisfaction among providers, patients, and families.3,9,56 A consensus-based approach can be used, including 1) the assessment of pain and sedation, 2) the designation of a target for “wakefulness” of the patient, and 3) the implementation of an algorithm to achieve the target with frequent reassessment.
Assessment of Pain and Sedation
Providers should use validated objective scoring for pain and sedation in young infants (Table 1). Those validated for full-term infants include the FLACC, CRIES, NIPS, and Objective Pain Score and are summarized in Supplemental Tables 4–7).57–59 The SBS is used to describe the level of sedation (from -3 unresponsive to +2 agitated) validated for the intubated patient not on neuromuscular blockade (NMB); the SBS target can be prescribed and documented (Supplemental Tables 4–8).6
Postoperative Pain Management
A multimodal approach to analgesia for patients extubated in the operating room or shortly after is recommended.60 In addition to opioids, intravenous acetaminophen as well as local anesthetics, adjunctive α-2 agonists (clonidine and dexmedetomidine), and N-methyl-D-aspartate antagonists (ketamine) can be used alone or in combination. Conversion to oral analgesics should occur after minimal enteral intake has been established. Nonpharmacologic therapies, including complementary and alternative medicine techniques with environmental control (soothing and bundling), distraction, and physical therapy, serve as important adjuncts to traditional pharmacologic approaches.61
Clinical Pathways
Standardized sedation/analgesia clinical pathways decrease variation in practice through guidance of initiation, titration, and weaning of continuous sedation/analgesia infusions and recommended alternatives to continuous infusions based on targeted sedation/analgesia levels.9,62–64 Patients with and without delayed sternal closure postoperatively not receiving NMB can be managed with clinical pathways as described below. Initial medications should be at the lowest dose to effect. Single-agent fentanyl offers analgesia as well as sedation for newborn patients.65 The α-agonist dexmedetomidine can be used in newborns at low doses and may offer an opioid-sparing effect.66 Initial sedation choices should attempt to avoid benzodiazepines to prevent potential neurodevelopmental side effects.67 Bedside care providers use validated scores for pain and sedation (Table 1) for the assessment of the adequacy of ordered medication treatments and communication to ordering providers. Patients receiving NMB for hemodynamic lability or optimization of metabolic demand cannot be assessed with objective scores and will not be eligible for clinical sedation/analgesia pathways. In addition to several studies outlining the benefit of standardized sedation/analgesia algorithms for the patient in decreasing total exposure to opioids and benzodiazepines, decreasing the incidence of withdrawal, and even revealing a decreased length of stay, both providers and families express increased satisfaction with pain control/sedation adequacy when these pathways are in place.7,8,56,64
The following pathways described below are illustrations of well-established and evidence-based protocols specific to 1 institution, Texas Children’s Hospital, Houston, Texas. They are identified as models to further establish a universal and standardized approach to sedation initiation, titration, and weaning in neonates with congenital heart disease.
Neonatal Cardiac ICU Sedation Pathway (Patients ≤30 days old) (Fig 1): Institutional protocol; Texas Children’s Hospital, Cardiac ICU, Houston, Texas
Age/weight-based pain and sedation initiation and management clinical pathway for neonatal cardiac patients.
Age/weight-based pain and sedation initiation and management clinical pathway for neonatal cardiac patients.
Fentanyl infusion as single agent with dexmedetomidine as secondary agent if required. Midazolam used only as an emergency rescue intermittent dosing or secondary infusion if dexmedetomidine is not hemodynamically tolerated.
If SBS is greater than desired, consider nonpharmacologic interventions to decrease environmental stress, including bundling (if possible), noise reduction, and soothing.
If SBS and/or pain score remains elevated, bolus doses with dosing matching 1 hour’s worth of infusion as a “rescue.” SBS/pain score reevaluated after each, as needed; dose given at 15-minute intervals to achieve prescribed sedation.
Incremental increases in infusion are suggested if 2 to 3 bolus doses are inadequate. The bedside nurse is instructed to notify the provider for infusion changes.
If SBS is less than prescribed, decrease the infusion to maintain the SBS and titrate to effect.
Tolerance and Withdrawal
Patients with prolonged narcotic or benzodiazepine exposure can develop tolerance (decreased response to the same doses of medications), dependence and withdrawal, hyperalgesia (heightened sensitization to pain), allodynia (pain elicited by nonpainful stimuli), or delirium.10,24,51,52
For patients at a high risk of iatrogenic withdrawal syndrome (IAWS), including those receiving a sedative infusion >5 days and particularly for those whose hemodynamics would not tolerate withdrawal symptoms (pulmonary hypertension, history of seizure disorder, depressed cardiac function, residual cardiac lesions after surgery, such as atrioventricular valve regurgitation, history of previous withdrawal syndrome), a weaning pathway is desired.51,68–70 Nonopioid drugs may also be considered (α agonists, ketamine, gabapentin) while weaning narcotics. Long-acting narcotics such as methadone should initially be avoided because of high cumulative and extended opioid exposure but may be necessary for certain patients anticipated to require a prolonged wean, such as those exposed to sedative infusions >21 days. The withdrawal assessment tool (WAT-1) (Fig 2) should be documented twice daily until 72 hours after the last habituation medication dose.70
Weaning Pathway (Fig 3): Institutional protocol; Texas Children’s Hospital, Cardiac ICU, Houston, Texas
Clinical weaning pathway for opioid and benzodiazepine with goal of wean duration to be less than total duration of intended usage.
Clinical weaning pathway for opioid and benzodiazepine with goal of wean duration to be less than total duration of intended usage.
Exposure ≤5 days: Opioid/ benzodiazepine can be stopped at the discretion of the medical team or weaned by 20% of the initial hourly rate Q8 to Q12 hours. As needed bolus doses are given to keep WAT-1 ≤4.
Exposure >5 days or high risk of IAWS: Opioid/benzodiazepine infusions weaned incrementally until oral thresholds are met (Fig 4). Note: Incremental changes in dosing should reflect a predefined percentage of day 0 dose (eg, 10%–20%).
Intermediate exposure 8 to 14 days: Wean 10% to 20% Q12 hours.
Long-term exposure >14 days: Wean 10% to 20% Q24 hours.
Chronic exposure >21 days: Consult a clinical pharmacist for weaning plan.
After oral conversion (Fig 4): Wait 24 hours for further weaning.
Pain Score ≥4 or WAT-1 ≥4: Intravenous rescue therapy used. If multiple rescues required per shift or hemodynamic status changes, hold further wean.
After lowest dose or “basement dose” reached (Fig 2), wean frequency daily until off.
Weight-based intravenous to enteral route dosing transition guidelines for opioids and benzodiazepines.
Weight-based intravenous to enteral route dosing transition guidelines for opioids and benzodiazepines.
Note: For patients on both an opioid and benzodiazepine, each agent should be weaned in alternating fashion.
In challenging patients with prolonged sedation complications from opioids and benzodiazepines, the pain service and pharmacy should be consulted and long-acting opioids, such as methadone, should be used.
Recommendations:
The use of standardized pathways guiding the initiation and titration of sedation and analgesia management based on objective, validated pain and sedation scores is recommended to maximize comfort while minimizing medication exposure (Class I; LOE A)
The implementation of a standardized weaning protocol is recommended for patients at a high risk of IAWS, including those receiving infusion(s) >5 days and particularly for those whose hemodynamics would not tolerate withdrawal symptoms (Class I; LOE B-NR)
Conversion to oral analgesics can be beneficial after minimal enteral intake is established (Class IIa; Level of Evidence C-EO)
A multimodal strategy of analgesia, including nonopioids, and nonpharmacologic interventions, such as integrated complementary and alternative medicine techniques with environmental control (soothing and bundling), distraction, and physical therapy, is reasonable (Class IIa; LOE C-EO)
Summary
Maximizing pain control and comfort while minimizing the side effects of sedative and analgesic medications is a critical component of the management of neonates after cardiac surgery. A multidisciplinary team-based approach using standardized pathways guided by validated scoring for pain and sedation to initiate, titrate, and wean sedative and analgesic medications while incorporating nonpharmacologic interventions optimizes this care, improving patient outcomes and satisfaction among providers and families.
Drs Smith-Parrish and Vargas contributed the section on preoperative assessment and sedation, including the tables and references; Dr Katherine Taylor contributed the section on intraoperative management, including references; Drs Achuff and Lasa contributed the section on postoperative analgesia and sedation, including tables and references; Dr Hopper planned, edited, and revised each section of the manuscript and combined them; Dr Ramamoorthy planned, edited, and revised each section of the manuscript, combined them, and wrote the abstract; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: The Neonatal Heart Society (NHS) contributed an educational grant to the project, Neonatal Cardiac Care Collaborative (NeoC3). The NHS, on a regular basis, applies and receives several unrestrictive educational grants for several internal projects from the following organizations and companies: Abbott Formula, Mead Johnson, Cheisi, Mallinckrodt, Prolacta, and Medtronic. The grants received from industry partners were used solely to offset the cost of publishing this supplement in Pediatrics. The industry supporters did not suggest manuscript content, nor did they participate in any way in the writing or editing of the manuscript.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest relevant to this article to disclose.
The guidelines/recommendations in this article are not American Academy of Pediatrics policy, and publication herein does not imply endorsement.
Comments