Short- and Long-term Impact of Neonatal Pain and Stress

After completing this article, readers should be able to: Pain is an inseparable part of everyday life. It is universal, protective, and crucial for survival. Pain can have profound deleterious and disruptive effects on the quality of life.—Menon, Anand, McIntosh, 1998Approximately 300,000 neonates are admitted to neonatal intensive care units (NICUs) annually in the United States. During their hospital stay, most infants undergo hundreds of diagnostic and therapeutic procedures that are necessary for their improved survival. Repetitive acute pain caused by invasive procedures, established pain resulting from neonatal diseases or surgery, and chronic/prolonged painful stimulation experienced by the NICU population often causes severe stress, probably leading to adverse neurologic outcomes. Follow-up studies of children and adolescents who were preterm infants have reported major developmental deficits, with lifetime needs for special assistance and increased health care costs for early intervention and special education. Early repetitive pain and maternal separation, both routine occurrences in the NICU, lead to permanent changes in peripheral, spinal cord, and supraspinal pain processing; neuroendocrine function; and neurologic development. These changes later manifest as abnormal pain states or pain thresholds, increased anxiety/stress disorders, attention-deficit disorder, or atypical behaviors such as hypervigilance and exaggerated startle responses. Although overall survival has improved dramatically, many NICU graduates have cerebral palsy, cognitive defects, attentional disorders, and other forms of long-term disability.Neonates are exposed to numerous painful procedures as well as chronic NICU care such as assisted ventilation. Large surveys from the United States and Canada have shown that neonates usually are treated for postoperative pain (70%), occasionally are treated for painful conditions such necrotizing enterocolitis (20%), and rarely are treated for routine procedures such as suctioning or intravenous access (<5%). Preterm neonates younger than 31 weeks’ gestation undergo the most procedures (74%) and are at greatest risk for developmental aberrations because abnormal sensory inputs occur at the time of active cortical (and peripheral) development. Because preterm neonates cannot complain or protest, it is imperative that caregivers understand and appreciate the consequences of neonatal pain so it can be treated appropriately. This review focuses on the neurodevelopmental aberrations occurring as a result of repetitive pain experienced in the NICU.Neonatal rat pups show exaggerated responses to nociceptive stimuli compared with older rats during development and adult rats. Maximal pain behaviors occurred in 50% of the tested rats with formalin concentrations of 0.52% for 3-day-old rat pups, 1.21% for 15-day-old rats, 4.56% for 25-day-old pups, and 5.60% for adult rats. Thus, the effects of inflammatory pain in 3-day-old rat pups were 11-fold greater than those in adult rats. Stimulation of the dorsal flexor reflex with Von Frey hairs also has demonstrated much lower thresholds for mechanical stimulation in newborn rats versus older rat pups. Using the tail-flick test, the pain thresholds for thermal stimulation were noted at 37.5°C, 46.5°C, and 47.2°C for rats at the ages of 3, 15, and 90 days (adults), respectively.Given the similar developmental events in preterm human and newborn rat brains, there is ample animal and human evidence for lower pain thresholds among preterm neonates or term neonates compared with older children or adults. Afferent stimulation in the early postnatal period produces long-lasting excitation of dorsal horn cells, and repetitive stimulation or tissue injury leads to increased excitability above and below the site of stimulation. Another study documented that preterm infants who received brief physical handling as if positioned for a lumbar puncture, then were allowed to rest for 10 minutes, showed increased physiologic responses and behavioral state changes from a subsequent heel stick than did nonhandled controls undergoing the same heel stick. The exaggerated reflexes, large receptive fields of the dorsal horn cells, and prolonged excitation after stimulation in rats mimics the low pain thresholds and hypersensitivity noted in preterm neonates. Thus, the combination of increased sensitivity to pain and greater plasticity of the developing brain make it vulnerable to long-term or even permanent alterations resulting from exposure to repetitive pain in early life.The physiologic effects of acute pain on the peripheral, spinal, and cortical nervous system are discussed in the article on the physiology of pain and stress in this issue and are reviewed only briefly here. Increasing degrees of tissue damage cause acute pain, which is followed by primary and secondary hyperalgesia that decreases pain thresholds in the area of and surrounding the damaged tissue. Repeated subthreshold (or innocuous) stimuli may elicit the windup phenomenon underlying primary and secondary hyperalgesia within the dermatomes supplied by spinal cord segments above and below the area of tissue damage, which increases the perception of pain in these areas. Preterm neonates younger than 34 weeks’ gestation lack the ability to modulate pain, progressively enhancing their perception of pain with greater immaturity. Early neuroanatomic development in the spinal cord and brainstem or cortical areas ensures the transmission and (perhaps) perception of pain, as evidenced by robust physiologic and behavioral changes. Thus, multiple lines of evidence reveal that newborns, who cannot verbalize their discomfort, may “feel” more pain.After acute painful stimuli, newborns show changes in various physiologic parameters, stress hormone levels, behavioral state, cry features, facial expressions, sleep states, behavior, and responses to developmentally supportive care. Autonomic responses lead to increased blood pressure, heart rate, respiratory rate, and palmar sweating; desaturation episodes; and changes in intracranial pressure. Blood pressure, heart rate, and respiratory rate changes have been documented during minor procedures such as heel lance, venipuncture, and circumcisions as well as during and after major surgery. Numerous studies have documented amelioration of these changes after local/topical anesthesia, sucrose, or other interventions. Palmar sweating increases in newborns undergoing painful procedures and has provided a useful research tool to evaluate the effects of analgesia or anesthesia for neonatal pain. All of these physiologic changes are a result of the autonomic changes associated with acute pain. Desaturation episodes and changes in intracranial pressure have been associated with acute pain in the newborn population. These episodes, often leading to hypoxia and subsequent hyperoxia, may cause cerebral blood flow changes and white matter damage in preterm infants. These changes may be important in the pathogenesis of intraventricular hemorrhage (IVH) and possibly periventricular leukomalacia in preterm neonates (Fig. 1). Thus, robust physiologic changes accompany painful stimuli in term and preterm neonates. These changes, which may increase the instability of sick neonates, also can be used to measure pain intensity in newborns.Stress hormone levels have been measured in newborns undergoing major surgery and some minor procedures such as venipuncture or circumcision. Anand and others have shown that neonates had significantly increased stress hormone levels, including cortisol, catecholamines, growth hormone, and glucagon, following surgery. Further, adequate anesthesia with inhalational and opioid agents significantly mitigated these effects. Elevated stress hormone levels in newborn rats were associated with increased neuronal excitotoxicity and adverse long-term neurologic outcomes. Most studies have shown little or no change in hormone levels with treatment for minor pain.Acute pain produces marked changes in behavioral state, crying, and facial expression. Infants exposed to repetitive pain exhibit sleeplessness and sudden or maladaptive behavioral state changes. In addition, they spend less time in the quiet alert state when they are able to absorb information from their environment or interact with their caregivers. The neonatal pain cry in most studies is different in terms of latency and duration, but crying decreases markedly in all neonates with decreasing gestational age, making it a poor indicator for pain. As neonates advance in gestational age, their disorganized or purposeless body movements are replaced with more coordinated movements, such as withdrawal in response to pain. In addition, preterm infants have less energy for movement with decreasing gestation, making body movements another poor indicator for pain. Interestingly, in a detailed observational study, postural, trunk, and limb movements were related to pain, but startles, twitches, and tremors were not. Many caregivers misinterpret the latter types of movements as indicators of pain.Neonatal pain and stress have been implicated as predisposing factors in the pathophysiology of IVH. For example, thrombotic complications and subsequent IVH are increased after surgery but are less common with adequate anesthesia. One pilot study also showed an increased risk of IVH among neonates not receiving opioid treatment for assisted ventilation. Intracranial pressure changes and cerebral blood flow changes using near-infrared spectroscopy and, more recently, Doppler measurements have been linked to IVH in numerous studies. These changes are associated with low perfusion/reperfusion injury in the brain, which is well documented in animals and humans after acute pain, although their precise role in the causation of IVH remains unclear. Thus, acute pain is associated with physiologic, hormonal, and behavioral changes that possibly could lead to IVH or other adverse outcomes.As noted in the article on the physiology of pain and stress, repetitive pain leaves a legacy of widespread changes in the neonatal brain that are accentuated in preterm neonates. Brain development and structure depend on results from neuronal replication, migration, differentiation, and interconnections related to function. After replication, regulation is achieved by cell-to-cell recognition events, which define the location, density, number, and morphology of the cellular elements in each brain area. After a rough definition of cellular elements, an exuberant production of dendrites, synapses, and synaptic vesicles occurs in excess of that needed for development. These patterns are refined by activity-dependent mechanisms matching axonal processes with target cells.Plasticity is the process of fine-tuning the neuronal development via activity-dependent mechanisms. Increased plasticity in the newborn brain permanently captures the effects of environmental inputs on development and, thus, affects future function and behavior. These effects are highest in the postnatal period, when neonates must interact with their extrauterine environment and incorporate their experiences into future function and development. Ample evidence shows that even the first few hours after birth are vitally important for maternal attachment and bonding in human and animal species.Neuronal activity results in the maturation, proliferation, and stabilization of various cortical processes; inactivity results in apoptosis. For example, if newborn kittens are exposed only to vertical lines, they trip over horizontally placed wires, unlike their normally raised counterparts. On the other hand, too much stimulation, which occurs with repetitive pain in the NICU, results in increased excitotoxicity that can lead to neuronal cell death and cause permanent alterations in development. Thus, cortical development is vulnerable to aberrations in postnatal activity and can be affected permanently by acute, chronic, or repetitive pain and stress.Multiple lines of evidence from animal and human studies reveal the long-term complications of neonatal pain and stress. Preterm infants are particularly vulnerable to repetitive pain and other stressors in the NICU environment because of their plasticity and their severity of illness. Although most infants weighing less than 1,500 g escape major developmental handicaps, many are left with “minor” deficits in cognition, learning disorders, attentional disorders, behavioral problems, and motor abnormalities. Deficits commonly seen in children who had been born preterm include impulsivity, inability to cope with novel situations, poor adaptive behavior, and specific learning deficits. When neonates born at 1,000 g who have normal intelligence are compared with term-born controls, there are differences in higher order executive functions of attention, planning, and problem solving. Sociodemographic factors often modify some but not all of these factors. For example, language, depressive tendencies, and intelligence quotient testing are related more to maternal factors, and visual-spatial abilities and anxiety are related more to neonatal factors. The acute changes in cerebral blood flow in preterm infants who have impaired autoregulation predispose neonates to IVH as well as neuronal cell loss and white matter injury. These aberrations are likely responsible for a wide range of subclinical brain alterations, including the learning and behavioral problems associated with extreme prematurity. Thus, preterm infants, especially those weighing less than 1,000 g, seldom escape unscathed and are at high risk for neurodevelopmental abnormalities (Fig. 2).Neonatal rat models have been used extensively to isolate the effects of repetitive pain on subsequent development. The neurologic maturity of newborn rats roughly corresponds to a gestational age of 24 weeks in humans, thus providing an extrauterine model for evaluating the specific effects of pain paradigms commonly found in preterm neonates in the NICU environment. Often, human neonates are not handled by their primary caregivers in the NICU because of hospital rules regarding visitation, inconvenience, or concern over safety. When neonatal rat pups were exposed to a similar lack of physical handling or maternal separation, they exhibited exaggerated hormonal and behavioral responses to air puff startle (emotional stress) and hemorrhage (physical stress). They mounted greater glucocorticoid responses that were associated with increased anxiety, alcohol preference, hypertension, immune compromise, and adverse long-term outcomes. Activation of the hypothalamic-pituitary axis during sensitive periods of development alters the gain on the adrenocortical system during the remainder of the rat’s life, leading to an increased susceptibility to stress and anxiety disorders, greater depression, and other problems. These effects also were noted in recent experiments in nonhuman primates.At the cortical level, repetitive pain results in increased neuronal cell death in several areas of the neonatal cortex. By evaluating the different cellular products of apoptosis (programmed cell death) versus those of excitatory cell death, the mechanisms of neuronal cell death following repetitive pain can be investigated in animal models. Interestingly, this effect may be mediated via N-methyl d-aspartate (NMDA) receptors because NMDA blockade using ketamine pretreatment ameliorates the excitatory cell death associated with repetitive inflammatory pain. Repetitive acute pain results in lower pain thresholds at postnatal days 16 and 22, increased alcohol preference, increased anxiety, defensive withdrawal behavior, and prolonged memory for chemosensory cues. In summary, animal evidence shows that repetitive pain similar to that associated with the NICU experience in human infants results in increased stress hormone levels and exaggerated responses at the peripheral, spinal cord, and cortical levels, especially in younger gestational age subjects.Research over the last 25 years demonstrates unequivocally that newborns are affected by repetitive pain and stress. Studies of pain in human preterm neonates usually correlate with animal model studies. The consequences of untreated pain, particularly in preterm neonates, may be the adverse neurologic outcomes noted previously, especially in infants of lower gestational age. Repeated heel sticks, one of the most common procedures performed in the NICU, result in hyperalgesic responses due to peripheral sensitization, exuberant outgrowth of nerve fibers supplying the injured area, increased expression of afferent nociceptors, and increased excitability of afferent nerve fibers. Central sensitization results from increased excitability of the spinal cord neurons. Topical anesthesia may alleviate some, but not all, of the adverse effects of repetitive heel sticks, which has worrisome clinical implications. Taddio and associates showed that term neonates circumcised without analgesia had exaggerated responses to immunizations at 4 to 6 months of age compared with treated controls. In contrast, Grunau has noted less sensitivity to commonplace pain at 18 months in former preterm infants exposed to repetitive pain, which was correlated with the length of NICU stay. Greater somatization also occurred in former preterm infants at 4 to 5 years of age. At 8 to 10 years of age, former preterm infants rated pictures of painful events, such as a child being hit with a baseball, as higher in intensity than their term-born counterparts. Anticipatory fear is noted by 6 months of age to immunizations and pin prick.Long-term memory of pain appears to be functioning even in preterm neonates. Long-term memory is possible only with the functional integrity of the hippocampus, limbic system, and other parts of the diencephalon. At a cellular level, the glutamatergic and opioidergic receptors and neurotransmitters as well as long-term potentiation and depression of cellular responses to environmental stimuli must be in place. All these systems are active in preterm neonates as early as 25 to 26 weeks’ gestation. Noxious stimuli probably are retained in the subconscious rather than the conscious memory. Nevertheless, it is clear that human neonates exposed to early postnatal pain show altered behavior to subsequent pain.Although there is extensive literature regarding the effects of pain on animal models and in vitro systems, there is relatively little follow-up data in former preterm neonates relating directly to pain during their NICU stay. There are three problems with conducting pain research in human neonates. First, there is an understandable reluctance due to ethical and humane concerns to have a pure placebo group when dealing with neonatal pain. In the recently completed NEOPAIN trial, 242 of 444 (55%) patients randomized to the placebo arm of the study received additional analgesia, which made interpretation difficult. Second, it is almost impossible to separate pain from other stressors in the NICU. Patients who are subjected to the most pain are the most immature and have the highest acuity, subjecting them to more stress in other areas. Third, the effects of repetitive pain are most likely to manifest in subtle neurologic aberrations that cannot be tested easily. For example, executive functions and the ability to solve new problems require long-term follow-up, which is time-consuming, expensive, and requires high follow-up rates that are difficult to achieve in the United States. Nevertheless, the potential difficulties associated with untreated repetitive pain mandate a firm research commitment.Novel ways to investigate pain in human neonates can provide meaningful answers to questions about controlling neonatal pain so as not to affect subsequent development. Different forms of pain relief, such as nonsteroidal anti-inflammatory drugs and other opioids besides morphine, should be evaluated. It is even possible that newer drugs can affect the cortical axis by modifying their effects on the developing neurons. of different could the for a placebo blood flow studies can be used to evaluate the effects of pain treatment on NICU patients in This to evaluate pain with from other must be to long-term studies that have an on higher order functions among former preterm neonates. This information is crucial to treatment of the pain that is necessary to survival in to pain both and Preterm infants are more vulnerable than their term-born because pain is in that because of their and because of the of multiple procedures necessary for their survival. and human data an role for a of acute and repetitive neonatal pain. is needed to the acute and long-term effects of pain on human neonates and to novel ways of