Over the past two decades, there has been significant progress in the understanding of neuroanatomy, physiology, and the pharmacology of analgesics in children, which has led to considerable advancements in pediatric pain management. This chapter discusses developmental anatomy and neurochemistry, pain assessment, pharmacologic treatment of pain, and techniques for providing pain relief.
The Infants and fetuses in the last trimester are neurologically sophisticated in their ability to transmit pain signals and respond to stress.1 Cutaneous sensory nerve terminals are present in the perioral region at 7 weeks’ gestation and spread to all body areas by 20 weeks’ gestation. Nerve growth factors regulate the extension of peripheral nociceptive fibers into the dorsal spinal cord, with the larger A fibers entering prior to the C fibers at 8 to 12 weeks’ gestation. At birth, A and C fiber territories overlap in the developing substantia gelatinosa.2 Therefore, the neonatal response to a nonspecific sensory stimulus is low-threshold, nonspecific, and poorly organized, as are the well-defined neonatal motor reflexes.3,4 Noxious and non-noxious stimuli produce similar physiologic and behavioral infant responses, which complicate an accurate assessment of pain.5
In the central nervous system (CNS), the full complement of cortical neurons, approximately 1000 million, is present at 20 weeks’ gestation. Pain transmission pathways complete myelination in the spine and brain stem between 22 and 30 weeks’ gestation. Myelination extends to the thalamus by 30 weeks, and to the cortex by 37 weeks or term. Research using near-infrared spectroscopy (NIRS) shows activation of somatosensory cortex in preterm infants after noxious stimulation.5,6 Cortical descending inhibition develops post-term. Excitatory and inhibitory neurotransmitters and neuromodulators are present in the fetus, with the balance favoring excitation. Calcitonin gene-related peptide (CGRP), substance P, and the glutamate-NMDA systems are present at 8 to 10 weeks’ gestation. Enkephalins and vasoactive intestinal peptide (VIP) appear at 10 to 14 weeks’ gestation. Catecholamines are present in late gestation, and serotonin at 6 weeks’ postnatal. Of note, the receptors for excitatory neurotransmitters are numerous and widely distributed in the neonate, regressing toward an adult system in the postnatal months. As well, in the developing nervous system, inhibitory chemicals, such as γ-aminobutyric acid (GABA) and glycine, may act as excitatory transmitters. In an experimental murine model, the spinal cord concentration of N-methyl-D-aspartate (NMDA) receptors, and their ligand-affinity, is greater in neonates than in older animals. Neurokinin 1 (NK1) receptor density is also maximal in late fetal and early postnatal life; however, substance P levels are lower at birth than adult levels.7
In relation to stress responses, the functional neuroendocrine pathways between hypothalamus and pituitary are present at 21 weeks’ gestation. Corticotropin-releasing factor (CRF) may stimulate fetal adrenocorticotropic hormone (ACTH) and β-endorphin from that time period, and cortisol and β-endorphin increases have been assayed following intrauterine sampling for exchange transfusion. Norepinephrine is present in paravertebral ganglia and adrenal chromaffin cells at 10 weeks’ gestation and is released with intrauterine stress (asphyxia). A smaller amount of epinephrine is present after 23 weeks’ gestation.7
The long-term consequences of untreated pain in the developing organism are currently being defined, and a number of studies suggest that early pain responses influence later pain behaviors.8 In one murine model, skin wounds on rat pups caused increased innervation and lowered pain thresholds in the area of injury for 3 months postinjury. In the rats that had recurrent painful stimuli from birth, the changes in the receptive fields of the dorsal horn neurons were persistent.9 Another study exposed rat pups to repeated hindpaw injections over several days. When compared with control pups and at adult age, the rats that experienced repeated noxious stimuli showed increased responses to painful and nonpainful stimuli relative to their controls. Pathologically, the experimental group showed a loss of nociceptive primary afferents.10 A third study found that the pain-conditioned behaviors of rats differed according to the timing of the stimulus. Rat pups exposed to early, repetitive noxious stimulation had a decrease in pain threshold compared with control rats. Adult rats that were given repetitive painful stimuli showed greater stress responses, such as freezing and digging, than controls.11
In humans, pain in infancy influences plasticity in a number of pain transmission pathways, including peripheral nerve sprouting, dorsal horn sensitization, decreased descending inhibitory control, and priming of the stress/hypothalamo-pituitary-adrenal (HPA) axis.12 Many but predominantly empiric studies show late behavioral effects of painful stimuli, and the balance of potential pharmacologic toxicity versus adequate pain control is an important consideration.13 One report describes that neonatal males who received a eutectic mixture of topical local anesthetic prior to circumcision had 12% to 25% less facial grimacing and tachycardia than randomized control infants without treatment.14
An earlier study reported that males circumcised by 2 days of age had longer periods of crying and higher pain ratings than uncircumcised males.15
Another report compared 18 preterm infants, subject to repeated painful procedures in the neonatal intensive care unit (NICU), with matched full-term infants regarding their somatic complaints at 18 months of age. Twenty-five percent of mothers of preterm infants with prolonged NICU stays noted a significantly increased number of somatic complaints in their toddlers compared with zero percent of mothers of full-term infants briefly managed in the normal nursery.16 Alternatively, a recent study of 24 preterm infants compared with matched full-term infants had similar behavioral pain scores when exposed to a finger prick at 4 months of age.17
In summary, the neonatal pain transmission system is adequately developed, centrally hyperexcitable, with the necessary components of central sensitization, but generally nonspecific in its response to stimuli. Neonates and infants feel pain, but assessment of the phenomenon remains challenging.18,19,20 Long-term effects of neonatal pain are being investigated; however data support prevention and management of pain in the newborn.21
Pain assessment is a fundamental and essential part of pain treatment. The ability to assess pain reliably facilitates the diagnosis of painful conditions and to evaluate efficacy of pain relief methods. The assessment of pain in infants and children, however, is one of the most difficult challenges faced by health care providers (HCPs) in part because of differences in verbal and cognitive developmental abilities, differences in pain expression and perception, and the subjective nature of pain. For these reasons pain assessment in infants and children requires a comprehensive approach that utilizes self-report when available, observations, and physiologic measures. As discussed, there is evidence to suggest that inadequately treated pain can have both short-term and long-term consequences. For example, it is well established that full-term and preterm infants develop physiologic stress responses to pain and inadequate anesthesia that can result in greater postoperative complications.1,13–17 Self-report methods are considered to be the most reliable guides to pain assessment for most patients. However, infants and preverbal children are unable to communicate their experience of pain and must rely on caregivers to interpret signs of pain and distress. Pain assessment methods that combine self-report with other measures, such as behavioral and physiologic responses, may provide more accurate measures of pain. There can be limitations in the use of behavioral and physiologic indices for pain assessment. The distinction between pain and distress may be difficult in young children. For example, a young child may cry and exhibit characteristic facial grimacing during an ear examination because of fear and anxiety rather than pain. Physiologic signs may also mislead measures of pain in certain situations. For example, patients who are septic, hypoxic, or receiving vasopressors may exhibit increase in heart rate or blood pressure that reflect other processes not related to pain. Most pain scales are designed for assessment of acute pain and tend to underestimate persistent or chronic pain in children.22
Most pain assessment scales used for infants and preverbal children rely on behavioral observation and physiologic parameters to guide assessment. Observational measures alone may not represent pain intensity accurately because HCPs tend to underestimate pain when compared with a patient or parent report.23,24 Parent report also tends to underestimate children’s pain but to a lesser extent than report by HCPs.25 Behavioral parameters typically used are facial grimacing, cry, body movement, and sleep pattern. The typical pain facial expression of eyes tightly closed, furrowed brows, and square mouth is considered to be one of the most consistent signals of pain in infants.26 Physiologic parameters such as heart rate, oxygen saturation, blood pressure, and palmar sweating provide objective evidence of pain.
The Premature Infant Pain Profile (PIPP) (Fig. 64-1) and CRIES (Fig. 64-2) are pain scales used for preterm and full-term infants, respectively, that combine behavioral observations and physiologic criteria to assess pain.27,28
FIGURE 64-1.
Premature infant pain profile.
The PIPP was specifically designed to assess acute pain in preterm infants with consideration to gestational age. The CRIES scale, an acronym for Crying, Requires O2, Increased vital signs, Expression, and Sleepless, consists of five behavioral and physiologic parameters designed to rate postoperative pain in neonates.
The FLACC scale combines five types of pain behaviors, including facial expression, leg movement, activity, cry, and consolability which have been shown to have good interrater reliability and validity in children (Fig. 64-3). It is widely used because it is quick, versatile, and can be applied to infants and older children, including those with developmental disabilities.29
Children aged 3 to 7 years become increasingly able to communicate their experience of pain to parents and caregivers. Children in this age group may not understand the abstract concept of pain, but most are able to indicate pain intensity using either pictographic or faces scales30,31,32 (Figs. 64-4, 64-5). Although self-report measures are most reliable, a number of factors may alter a child’s report of pain.33 For example, children with inadequately treated persistent pain from cancer or surgery may appear very quiet, still, and withdrawn, giving a false impression of adequate analgesia. Some children may underreport or deny pain for fear of receiving a painful analgesic “shot.” Numeric scales are not useful in this age group because although many are proficient at counting, children younger than 7 years do not understand the quantitative significance of numbers. Several self-report methods have been developed that are validated and reliable in children as young as 4 years of age.34,35,36 The Bieri faces scale is a series of facial expressions depicting degrees of pain and was found to be preferred by most children.37,38
Children aged 8 years and older generally can use standard “0 to 10” visual analog scales accurately, but many of the scales used in younger children such as the Bieri faces can also be used. Children in this age group may have concerns over loss of control, or may fear painful injections of analgesics that can distort their self-report. Older children and adolescents have the cognitive ability to understand the meaning of pain and tend to use behavioral coping strategies for pain.
Pain assessment in children with developmental disabilities is particularly challenging for parents and caregivers. Pain assessment in this population of children must be based on the child’s individual abilities. Some children with developmental disabilities can self-report and should be given this opportunity. Others require behavioral or physiologic measures of pain. In the last decade, several pain assessment tools have been developed addressing the specific needs for this patient population.39,40,41
In general, pain assessment is best accomplished by correlation of self-report, behavioral, and physiologic measures with the child’s overall clinical picture. The choice of pain assessment scale should be individualized and based on a child’s age, clinical condition, environment, cognitive abilities, and coping style. Often, explanation and practice of the chosen pain assessment scale during the preoperative visit can facilitate the child’s use after surgery.
Infants and young children have age-related differences in the pharmacokinetics and pharmacodynamics of analgesics which are relevant to the safe and effective dosing of analgesics in this age group. Analgesics with high-water solubility have a larger volume of distribution, sometimes resulting in the need for larger initial dosing. However, neonates and young infants have immature hepatic enzyme systems involved in conjugation, glucuronidation, and sulfation of analgesics such as opioids and amide local anesthetics, which cause prolongation of elimination half-life and increase the risk of drug accumulation. Most infants and young children will have maturation of hepatic enzyme systems by 6 months of age; however, there is considerable variation in maturation rates. Neonates have decreased plasma-protein binding due to decreased levels of albumin and α1 acid glycoprotein, resulting in increased free pharmacologically active drug and greater first-pass toxicity. Renal function, including glomerular filtration and renal tubular secretion, is decreased in the first few weeks of life compared with adults. Renal immaturity also results in the slower elimination of glucuronides of morphine, hydromorphone, and of monoethylglycinexylidide (MEGX), a principal metabolite of lidocaine. In addition, infants, particularly premature infants have immature ventilatory reflexes in response to hypoxia and hypercarbia and have increased risk of hypoventilation in response to opioids. Because of the developmental pharmacokinetic and pharmacodynamics differences in neonates and young children, dosing of opioids and local anesthetics require careful titration and increased vigilance for side effects.
Non-opioid analgesics include acetaminophen, nonsteroidal anti-inflammatory drugs (NSAIDs), and selective cyclooxygenase (COX-2) inhibitors. Analgesic effect occurs from peripheral and central actions involving inflammatory processes in the spinal cord and brain.42,43 Because of their opioid-sparing effect, non-opioid analgesics are often first-line treatment for mild to moderate pain in children.
Acetaminophen is the most commonly and widely used analgesic in children and has a good safety record in children of all ages.44 The analgesic and antipyretic effects of acetaminophen are largely via sites within the CNS through action at cyclooxygenase (COX-3 and COX-2) isoenzymes, cannabinoid receptors, and on tyrosine-related protein (TRPV1) receptors.45,46 Although a weak analgesic, acetaminophen is a useful adjuvant for acute pain treatment and is often combined, synergistically, with opioids. The elimination of acetaminophen is primarily through glucuronidation and sulfation; elimination rates are similar among infants, children, and adults.47 The recommended single dose is 15 to 20 mg/kg and 10 to 15 mg/kg with repeated dosing. Maximum daily dosing is 100 mg/kg/day in children, 75 mg/kg/day in term infants, and 40 mg/kg/day in preterm infants. Inadvertent overdosing can lead to fulminant hepatic failure.48,49 Acetaminophen is available in several routes of administration—intravenous (IV), tablets, capsules, suspensions, and suppositories. Concentrated infant drops have been discontinued to reduce inadvertent overdosing.50,51 The IV dosing in children 2 to 12 years is 15 mg/kg every 6 hours or 12.5 mg/kg every 4 hours with a maximum daily dose of 75 mg/kg/day. The rectal dose is 35 to 40 mg/kg initially followed by 20 mg/kg every 6 to 8 hours; absorption is low and variable. Rectal absorption peaks at 70 minutes.52,53
Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used for mild-to-moderate pain and are often combined with opioids to improve analgesia and to help reduce opioid side effects. The use of NSAIDs has been shown to reduce postoperative opioid use by approximately 30% to 40%.54 The anti-inflammatory effect of NSAIDs is through reversible inhibition of COX-1 and COX-2 isoenzymes and inhibition of the conversion of arachidonic acid to prostanoids.55 The clearance of ibuprofen, ketorolac, and several other NSAIDs is more rapid in toddlers and preschool children compared with adults.56 NSAIDs in general have a good safety margin in children 6 months of age and older, particularly with short-term use. A large-scale study in children administered short-term use of ibuprofen showed a very low overall risk of severe side effects.57 There are limited safety data on the use of NSAIDs in neonates and young infants.58,59 Much of the pharmacokinetic and safety data of NSAIDs in use in neonates come from the use of ibuprofen and indomethacin to facilitate closure of patent ductus arteriosus. Ibuprofen is associated with less risk of renal toxicity and hyponatremia compared with indomethacin in this age group. Significant bleeding caused by NSAIDs is relatively uncommon in healthy children. There are mixed data regarding the risk of NSAIDs use in children after tonsillectomy procedures.60–62 Although analgesic trials show good analgesia, due to the potential risk of life-threatening bleeding after tonsillectomy procedures, the practice at this institution is to avoid NSAIDs in the perioperative period for children undergoing tonsillectomies. There is evidence in adult patients to suggest that NSAID use can impair bone healing after orthopedic surgeries that involve osteoclast activation and new bone formation. Children are less likely to have impairment of new bone formation; however, it is reasonable to avoid use of NSAIDs in children after orthopedic surgeries requiring significant active bone formation. Table 64-1 lists dosing of commonly used non-opioid analgesics.
Dosing Guidelines for Non-Opioid Analgesicsa
Dose < 60 kg | Dose > 60 kg | |
Acetaminophen | 10–12.5 mg/kg q4h PO or IV | 650 mg q4h PO or IV |
Ibuprofen | 6–10 mg/kg q6–8h PO | 400–600 mg q6h PO |
Naproxen | 5 mg/kg q12h PO | 250–500 mg q12h PO |
Celecoxib | 2–4 mg/kg q12h PO | 100–200 mg q12h PO |
Ketorolac | 0.5 mg/kg q6–8h IV, not for > 5 d | 30 mg q6–8h IV, not for > 5 d |
Opioids are widely used for the treatment of infants and children with moderate-to-severe pain. Safe and effective administrations of opioids require careful patient selection, an understanding of age-related differences in metabolism, dose titration, and aggressive treatment of opioid side effects.
As discussed, opioids are associated with an increased risk of respiratory depression and apnea and a prolonged duration of action in neonates and infants, particularly in premature infants, because of delayed hepatic enzyme maturation and immature renal excretion. Neonates and young infants also have reduced protein binding as a result of developmental changes in the expression of P-glycoproteins in the gastrointestinal tract and in the blood-brain barrier. Data from infant rat models show immature opioid receptor sites in the peri-aqueductal gray matter and descending pathways. Studies of use of opioids for procedural pain in neonates have shown mixed results; randomized trials assigning ventilated neonates to receive morphine infusions versus placebo infusions have not shown clear advantages in the morphine infusion groups.63,64 In addition to neonates and young infants, others at risk for opioid-induced respiratory depression include children with obstructive sleep apnea, children with craniofacial abnormalities, and children with neurologic conditions. Dose reduction with careful dose titration, cardiorespiratory electronic monitoring, and close observation are necessary for safe and effective opioid treatment in neonates, infants, and children at risk for respiratory side effects.
Evidence indicates that codeine is ineffective as an analgesic and is associated with significant side effects.65–68 Codeine is a pro-drug and is converted to morphine through the CYP2D6 enzyme pathway. There is significant polymorphism of CYP2D6 enzymes, leading to patients who are poor metabolizers, rapid metabolizers, and ultra-rapid metabolizers.69,70 Pharmacogenic data show that 30% of children are poor metabolizers and unable to convert codeine to morphine, making codeine inactive in these patients.71 Gene duplication in ultra-rapid metabolizers increases the amount of morphine metabolized from codeine. There have been fatalities and life-threatening events due to opioid overdose among children who were ultra-rapid metabolizers.70,72 Breast-fed infants are at risk of overdose when their lactating mothers, having been prescribed codeine-containing products for postpartum analgesia, are ultrarapid metabolizers.73 As a result of the safety and efficacy concerns for codeine, our institution has removed codeine from its formulary and from all medical record prescription software.74,75
Oxycodone is widely used in children with moderate pain, particularly in the postoperative setting when transitioning from IV opioids to oral opioids. It is metabolized via CYP3A4 to inactive metabolites, although a secondary pathway involving CYP2D6 generates potent oxymorphone, which is renally eliminated and has been associated with increased risks among ultrarapid metabolizers.76,77 Pharmacokinetic data of oxycodone in children show significant variability in clearance and elimination half-life, particularly in neonates.78 Our prevailing impression is that oxycodone is associated with fewer side effects in children compared with codeine. Typical starting doses are 0.05 to 0.1 mg/kg every 4 hours as needed for mild pain and 0.1 to 0.2 mg/kg every 4 hours as needed for moderate to severe pain. Oxycodone is available in an elixir form for children unable to swallow pills. Sustained-release preparation of oxycodone (OxyContin) is used for older children with chronic pain requiring opioids.
Morphine is often the first-line opioid considered for parenteral use in children. There is extensive pharmacokinetic data for morphine in children of all ages.79–82 Metabolism occurs primarily in the liver via glucuronidation to morphine-3-glucuronide which has neuroexcitatory properties such as delirium, myoclonus, and agitation; and to morphine-6-glucuronide which has analgesic, sedative, and respiratory depressant actions. Glucuronides are renally eliminated and can accumulate in children with renal failure. Data suggest that morphine is metabolized preferentially to morphine-3-glucuronide in neonates, increasing the potential risk of seizures in this age group. Elimination half-life of morphine is prolonged in infants and neonates, particularly in premature infants whereas clearance of morphine is reduced, increasing the risk of opioid side effects.82,83