BEHAVIOR, PAIN PERCEPTION, AND THE EXTREMELY LOW-BIRTH WEIGHT SURVIVOR

OUTCOME OF THE VERY LOW-BIRTH WEIGHT INFANT

0095-5108/00 $15.00

+ .OO

BEHAVIOR, PAIN PERCEPTION, AND THE EXTREMELY LOW-BIRTH WEIGHT SURVIVOR Michael F. Whitfield, MD, and Ruth Eckstein Grunau, PhD

The systematic application of neonatal intensive care as we now know it began, at least in tertiary centers in the Western world, in the early 1970s, and resulted in a dramatic improvement in survival rates and reduction in handicap rates for premature infants. This neonatal intensive care technology was an offshoot of in vivo physiologic studies on immature animals that had provided better understanding of the physiology of the human newborn, and the physiologic foundation for life-saving intensive care management. Neonatal intensive care aimed to correct life-threatening and handicapping problems related to temperature regulation, acid-base balance, and nutrition. A further important development was improvement in laboratory equipment for measurement of blood gases and other critical parameters on small blood samples, making repeated measurements feasible in small patients. In addition, some critical refinements in incubators, ventilators, and ancillary equipment designed for management of respiratory problems, bedside measurement of inspired oxygen in infants, and functional cardiorespiratory monitors first became available. Improvements in plastic technology provided small tubes of relatively nontoxic material for feeding tubes, endotracheal tubes, umbilical catheters, chest drains, and other vascular access devices. The introduction of this technology was exciting and provided for the first time the opportunity for effective clinical manageFrom the Department of Paediatrics, University of British Columbia, B.C.’s Children’s Hospital (MFW); and Centre for Community Child Health and Health Evaluation Research, BC Research Institute for Children’s and Women’s Health (REG), Vancouver, Canada

CLINICS IN PERINATOLOGY VOLUME 27 NUMBER 2 JUNE 2000

363

364

WHITFIELD & GRUNAU

ment of previously fatal or life-threatening illness in small and sick babies and produced dramatic results. It did, however, bring with it the need for an escalating armamentarium of invasive procedures that have become the normal experience of neonatal intensive care; witnessed and practiced by us, the caregivers, and experienced personally on a day-today basis by our small patients. These range from apparently innocuous procedures, such as diaper changing, to obviously noxious procedures including endotracheal intubation, ventilation, repeated blood sampling, and many other procedures involving direct tissue damage. Before full-scale neonatal intensive care, baby management had involved feeding; temperature control; occasional measurements of blood sugar, electrolytes, or hematology parameters; investigation and treatment of infections; intravenous fluids for those who were not taking adequate fluids or were hypoglycemic; circumcision for some; and the heelprick Guthrie screening blood sample for all before discharge. There was no doubt in the mind of any cognizant onlooker, even at that stage of baby care, that even babies down to 1000 g responded to the relatively small number of procedures that are painful in adults in a way that suggested that they experienced pain. Those of us swept along in the enthusiasm for the new neonatal intensive care tried to reduce the number of painful events while trying not to compromise quality of physiologic care. Babies were usually given opiate analgesia postoperatively following an albeit conservative regime. We thought or hoped that in some way the babies would not remember the procedural pain because of immaturity of the central nervous system. We used local anesthetic for local procedures where time permitted, which it frequently did not. We were desperately concerned about the potential side effects of effective analgesics, such as opioids. In particular, we worried about respiratory depression in immature babies who at the best of times had a definite tendency to apnea. In the nursery environment of those days the nursing staffing ratios were very low by today's standards and frequently all of the small number of baby ventilators available to the nursery were already in use. An additional unanticipated apneic baby resulting from opiate administration presented a major problem. Anand's et a18 small study of consequences of anesthetic practice during ligation of the patent ductus arteriosus was of fundamental importance in raising the clinical issue of the consequences of unacknowledged pain and stress in premature babies. In this randomized study, 16 preterm babies underwent patent ductus arteriosus ligation. In both groups nitrous oxide and paralysis were used for anesthesia, but in one group fentanyl (a potent opioid analgesic) was given intraoperatively in addition. Not only were measures of the hormonal response to the stress of surgery greater in the group who received no intraoperative opioid, but those patients also had a greater catabolic response in the days after surgery, and had more circulatory and respiratory complica-

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

365

tions postoperatively. This study clearly established that premature babies do mount a stress response to a significant intrathoracic operation despite narcosis with nitrous oxide and paralysis, and that this can be muted by the additional administration of a potent analgesic. Similar results were obtained in a further study with h a l ~ t h a n eEpinephrine, .~ norepinephrine, cortisol, and insulin-glucagon ratio correlated with severity of surgical tress.^ Two contemporary reviews further discussed surgical pain and stress7,92 and the beginning of recognition of differences between term and preterm infants? The presence of an apparently functionally intact neuroendocrine pain system raised many questions of direct clinical significance requiring investigation.6,15, 20, 26, 49, In addition to hormonal responses, the behavioral and autonomic dimensions of pain responses needed to be investigated and quantified. What differences might there be in quality and amplitude of response between term and preterm babies? How might they be influenced by gestational and postnatal age? Might the pain responses be altered by repeated noxious events? Might the responses and consequences induced by chronic pain be different from those of acute procedural pain? Might there be long-term consequences of repeated noxious events extending into childhood and perhaps beyond? Conversely, what might be the long-term consequences of administration of potent opioids on the developing brain?

ACUTE CONSEQUENCES OF PAIN IN THE NEONATAL PERIOD

Physiologic responses to painful stimuli include acute increases in heart rate, blood pressure, variability in heart rate, and changes in autonomic tone. Venous and intracranial pressure may rise, arterial oxygen saturation and skin blood flow fall.'O Alterations in intracranial blood volume and cerebral blood flow result from changes in intrathoracic pressure and respiratory movements associated with acute pain.65 Invasive procedures, such as tracheal suctioning, feeding tube insertion, or heel prick blood sampling, may be associated with marked changes 98 Frequent occurrence of in cerebral blood flow and oxygen delivery.70* these events in the early days of extrauterine life are associated with an increased risk of development of significant intraventricular hemorrhage or periventricular leukomalacia, which, in turn, are associated with an increased risk of adverse neurodevelopmental ~ e q u e l a e There . ~ ~ is a pathway defined by which an individual's acute neonatal pain experience could influence the occurrence of subsequent major handicap. Neonatal pain experience could also modify the development of the central nervous system in more subtle ways contributing to less overt neurodevelopmental and behavioral difficulties described in later childhood, as discussed later.

366

WHITFIELD & GRUNAU

THE PROBLEM OF IDENTIFICATION OF PAIN IN NEWBORNS

Considerable research in pain assessment of neonates during this decade has resulted in the development of many measures for assessment of acute procedural pain in full-term and preterm infants.’, 27 Changes in facial activity, together with shifts in infant sleep-waking state, and physiologic indices of heart rate are the most widely used pain indicators in preterm infants, with the addition of cry with termborn infants. Cry is not vocalized under mechanical ventilation, and is precluded as a pain cue for the smaller and sicker infants in neonatal intensive care. Observation of body movements is generally precluded when infants are bundled, as is the usual practice in developmental care in the neonatal intensive care unit (NICU). Of the large number of infant pain scales, none has been validated sufficiently for clinical use with infants less than or equal to 28 weeks gestational age. Multidimensional assessment of pain measuring both behavioral and autonomic aspects is essential because each potentially yields different information. This is approached in two ways. Composite scales additively combine elements of physiologic and behavioral responses, providing a summary score. The shortcoming of this method is that the weightings given to each component of the scale are essentially arbitrary. Composite scales include the Premature Infant Pain Profile,s6the Neonatal Infant Pain Scale,53and the COMFORT scale.3 The alternative to composite scales is to measure multiple response parameters, but not to combine them into a single raw score. The advantage of the latter approach is that it allows for study of associations between behavioral and physiologic systems, because relationships between responses may differ across gestational age and vary across situations. The most widely used univariate scale is the Neonatal Facial Coding System (NFCS).35,36 The NFCS is an anatomically based measure in which 10 discrete facial actions are coded as occurring or not occurring moment to moment. These facial actions are brow lower, eye squeeze, nasolabial furrow, open mouth, vertical stretch mouth, horizontal stretch mouth, chin quiver, lip purse, taut tongue, and tongue protrusion. Construct validity was established with term-born and preterm infants. NFCS facial actions discriminate tissue insult (painful) and nontissue insult (stressful but not painful) proceduresz0,33, 41,48, 78, 85 and differentiate infants receiving 83 or sucrose71during invasive procedures. Convergent validity 0pioids~~7 has been demonstrated through comparison with comprehensive facial coding.19Whereas the NFCS was developed for stop frame slow-motion video coding, it was recently adapted for bedside use.33Currently in our center for research purposes we use the NFCS together with measures of heart rate variability in infant pain studies.” Distress is displayed behaviorally and physiologically as a continuum at all ages, with varying degrees of pain at the upper end. Although pain scales have demonstrated validity in discriminating acute procedural pain, none of the cues (behavioral, physiologic, or hormonal)

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

367

are solely specific to pain. In nonverbal populations, behavioral and autonomic changes indicate distress of varying intensity, which then requires the caregiver to check for the source, often by a process of elimination. Surprisingly, when adult judges rated degree of pain in infants they observed on videotape, they attended more to facial display than fundamental frequency of the cry in studies of term-born18 and preterm infants.45 SHORT-TERM SEQUELAE OF PRIOR EXPOSURE TO NOXIOUS STIMULI

There are consistent data that preterm infants have a lower threshold to tactile stimulationz6,74 and appear to be hypersensitive to painz5 In a heel used for repeated blood sampling, preterm infants born at 27 to 32 weeks' gestation had a threshold for the flexor withdrawal reflex on that side that was half the threshold of the intact heel. This hypersensitivity caused by tissue damage is analogous to the tenderness reported in adults. Treatment of the damaged area with the topical anesthetic EMLA reversed this hypersensitivity, whereas placebo had no effect. These studies show that the premature infant is capable of mounting a chronic pain response to local injury that can be reduced by local anesthesia. In addition, not only do premature infants have increased sensitization to invasive procedures but also show significant overt physiologic and biobehavioral responses to repeated apparently nonnoxious tactile and nontactile stimuli and noise including changes in oxygenation and cerebral blood flow.28,63, 97 A further current and very worrying concern is the concept of the wind-up phenomenon following repeated painful events resulting in normally nonnoxious stimuli being perceived as pain with cumulative deleterious effects.", Extremely low-birth weight (ELBW) infants exposed to repeated and prolonged procedural pain in the neonatal period display a decreased behavioral response to a further acute painful event at 32 weeks' postconceptional age, less response being correlated with a greater number of prior invasive procedures. Johnston and Stevens48presented the first direct evidence linking prior pain events to pain observed subsequently. Infants were studied at 32 weeks' postconceptional age experiencing a heel lance blood collection. Preterm infants born within 4 days of the blood sampling observed were compared with infants of the same postconceptional age born 4 weeks earlier (at a mean gestational age of 28 weeks) undergoing the same procedure. This second group of infants had spent the period from 28 to 32 weeks' postconceptional age in a NICU and had undergone an average of 71 procedures compared with 12 in the first group. The earlier born infants had less behavioral manifestations of pain in response to heel lancing but had higher maximum heart rates and lower oxygen saturation than the newly born infants before and during the procedure. These physiologic and behavioral differences in pain response were interpreted as being related to spend-

368

WHITFIELD & GRUNAU

ing postconceptional weeks 28 to 32 in an NICU setting. Data from our own center supports the linking of the number of prior pain procedures with subsequent pain and stress behaviors. In 136 babies undergoing heel lance for blood collection at 32 weeks’ postconceptional age, the total number of invasive procedures from birth was related to diminished NFCS facial activity and infant behavioral state (Grunau et al, unpublished data). In contrast to the findings of Johnston and however, power spectral analysis of heart rate variability in the latter study (rather than maximum heart rate) showed autonomic and behavioral effects in the same direction. In another study among infants born less than or equal to 1000 g, the number of invasive procedures in the preceding 24 hours was related to increased brow raising stress behavior observed during intrusive handling of endotracheal suctioning, chest physical therapy or diaper change.32There is a paucity of data of the effects of chronic pain, either in the short- or long-term effects on subsequent pain responses. To our knowledge there is only one study in which ELBW infants have been directly systematically observed and compared with termborn controls while undergoing an invasive procedure after the neonatal period.@ In this study behavioral and autonomic indices of pain responses were compared following finger lance for blood sampling at 4 months corrected chronologic age (CCA) in ELBW infants and termborn controls aged 4 months. This study was an attempt to provide direct objective evidence of the apparent pain insensitivity reported by parents in 18-month ELBW children discussed later.40The results of this study provided detailed description of pain responses of 4-month-old controls and ELBW infants in a behavioral laboratory setting, and showed subtle measurable differences potentially attributable to prior pain experience in the ELBW group. A major and important difference in the current literature of the effect of prior pain experience appears to be the direction of the effect in healthy term and preterm infants: prior pain experience in healthy term babies appears to increase subsequent behavioral response to pain,88,89 whereas in prematures prior pain experience appears to diminish subsequent behavioral response to pain (Grunau et al, unpublished 64 The healthy term and preterm infants have had very different data).40, prior pain experience, however, in current studies forming the basis for this evidence. A healthy term baby may have experienced one to three noxious procedures in the neonatal period that might include neonatal circumcision without anesthetic. By contrast, neonatal intensive care involves 2 to 10 invasive procedures per day between 28 and 32 weeks’ postconceptional age,48or cumulatively 488 procedures during the NICU stay for a baby of 23 weeks’ gestation.I6In addition, other factors may constitute causes of ongoing discomfort or stress including causes of chronic pain, such as necrotizing enterocolitis, noise, and other apparently innocuous events in the normal nursing care of the baby that may contribute to perceived pain because of wind-up. The more procedures the baby has experienced, the less the baby tends to provide recognizable

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

369

behavioral cues to the caregivers that there is real distress (Grunau et al, unpublished data).&With this in mind, it might seem unlikely that there would not be long-term changes in a preterm infant who had lived through such an experience early in development. LONGER-TERM SEQUELAE OF PRIOR EXPOSURE TO NOXIOUS STIMULI

To date, there are very few longitudinal studies that have examined the effects of early pain on subsequent pain responses and attitudes. This is an area of significance for healthy term-born children undergoing routine procedures and of considerable potential significance for children who survive prolonged periods of neonatal intensive care. In term babies, neonatal circumcision provides an opportunity for observing later effects of a single potentially painful procedure. A retrospective study initially reported that full-term healthy circumcized infants showed a stronger pain response to subsequent routine vaccination at 4 to 6 months compared with the response in uncircumcized boys.88This was followed by a placebo-controlled study in which circumcized infants showed a stronger pain response on subsequent routine vaccination than uncircumcized infants.89In the circumcized group, preoperative treatment with the topical anesthetic EMLA (a eutectic mixture of lidocaine and prilocaine) attenuated some of the pain response to vaccination. The differences at 4 months among the circumcized infants were limited to one of three pain measures (observer ratings), however, whereas no differences were evident on either cry or facial activity, which are generally more reliable pain indices than global ratings.31 Pain sensitivity by parent report at 18-months corrected age was investigated in two groups of ELBW toddlers (< 801 g and 801 to 1000 g) compared with two control groups, one of heavier premature children and a group of full-birth weight children.40This study was prompted by parents of ELBW 18-month toddlers (born during the late 1980s) noting apparent lack of response to the normal bumps and hurts of toddlerhood and asking us in the clinic how to teach their toddler about pain. Both groups of ELBW toddlers were rated by parents as significantly lower in pain sensitivity compared with both control groups. Child temperament was strongly related to rated pain sensitivity in the full-birth weight gfoup (as would normally be expected) but was progressively less related to reported pain sensitivity in groups of decreasing birth weight, so that in the children of less than 801 g temperament and reported pain sensitivity were apparently independent of one another. Parenting style, a variable that might have explained some of these differences, did not mediate ratings of pain sensitivity. Number of prior neonatal painful experiences were not quantified in this study. This recognition of apparent lack of normal pain sensitivity in toddlerhood in ELBW children may be of clinical significance to emer-

370

WHITFIELD & GRUNAU

gency room management of ELBW children. Not only are gross motor co-ordination and motor planning less well developed than in full-birth weight toddlers, but the apparent response to the normal pain consequences of significant injury, such as a fracture following a fall, may be muted, making recognition of the severity of the injury difficult. We know anecdotally of several of our ELBW toddlers who have been sent home from emergency rooms with significant unrecognized fractures apparently because of lack of behavior suggesting a significant injury. Possible long-term effects of early pain experience in children aged 4.5 years was investigated in a further study looking at factors influencing somatization in children of less than 1001 g birth weight compared with socially comparable full-term ELBW children had significantly higher scores at 4.5 years for somatization (nonspecific physical complaints of no known medical cause) by parental report. Only ELBW children had scores in the clinically significant range. Factors related to somatization are complex. At age 4.5 years child temperament and parenting factors appeared to be contributory. The relationship of family dynamics and parenting characteristics to pain behavior in children in general are ~nclear,2~, 47, 62 and there are no other data we have found related to vulnerable ELBW children. Bodily pain of unknown cause is a relatively common complaint at school age and may be triggered or experienced differently in children with varying pain histories, but this question has not yet been studied. The limitation of these studies, which have attempted to examine possible long-term effects of pain in former ELBW infants, are that they were dependent on parent report. As a first attempt to compare the responses of the children rather than their parents, children’s judgments about pain in mid childhood were studied comparing children of less than or equal to 1000 g birth weight at 8 to 10 years with full-birth weight peers.%In this study children were shown and asked to rate the 24 pictures of the Pediatric Pain 1nvento1-y~~ using the Color Analog Scale and the Facial Affective Scale.60The 24 pictures are line drawings that depict painful events in four settings: (1) medical, (2) recreational, (3) daily living, and (4) psychosocial. Although the two groups of children did not differ overall in their perceptions of pain intensity or affect, the ELBW children rated medical pain higher than psychosocial pain, unlike the full-birth weight group. Pictures of pain during recreation (e.g., being hit by a baseball, getting hurt in a football game) were rated higher in pain intensity by the ELBW children. Duration of neonatal intensive care hospitalization for the ELBW children was related to increased affective discomfort of pictures of pain in recreational and daily living settings. These subtle differences between the two groups contrasted with the stark differences in apparent pain sensitivity by parent report between ELBW and larger-birth weight children at 18 months.4O Parent report of somatization was obtained again at 8 to 10 years, but did not differ at this age.34It is difficult to know whether the differences between the 8- to 10-year-old ELBW children’s responses to

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

371

pictures were caused by prior pain experience. There are other potential contributory factors to these findings that must be kept in mind. ELBW children are less well coordinated at this age and may have experienced more bumps and hurts in recreational settings than their term-born peers, which could have affected their ratings of recreational pain. ARE PRIOR PAIN AND LATER NEURODEVELOPMENTAL AND BEHAVIORAL OUTCOME CONNECTED?

The neurodevelopmental outcomes of expremature children have been extensively documented, including the smaller sicker babies at the lower end of the birth weight and gestational age spectrum. Apart from children with severe neurologic and sensory impairment, a whole spectrum of developmental problems are more prevalent among expreterm infants than peers. Sequelae include deficits in cognition, learning disorders, poor motor performance, behavior problems, and attention.* Premature very low-birth weight children have been found to have more educational, psychological, behavioral, and emotional difficulties than their full-birth weight peers. These include lower intelligence quotient (IQ) scores, problems with visual memory and visual perception, attention deficit disorder, hyperactivity, depression, anxiety, and low self-esteem at school age43, 95 persisting into adole~cence?~, 55 We examined functional abilities at age 8 to 9 years in 115 children of birth weight 800 g or less born between 1974 and 1987 and compared them with 50 peers of comparable age, sex, and social class.95Severe multiple disability was present in only 14%. A further 13%had borderline intelligence. ELBW children with global IQ more than 84 also scored significantly lower in standardized tests of fine and gross motor coordination; visual-motor pencil output; visual memory; and academic achievement (reading, written language, and arithmetic). Learning disorders occurred in 47% of the ELBW children with global IQ more than 84 compared with 18% of the controls, and 22 (41%)of 54 of the ELBW children who had learning disabilities had multiple areas of learning difficulty. ELBW children also had significantly worse scores on ratings of behavior during testing by the psychologist and by parental report. ELBW children were rated significantly lower than their peers in relation to dimensions of attention, activity, independence, problem solving, examiner support, and rapport on the Stanford Binet Behavior Rating Scale. In particular, ELBW children tended to be more active and anxious, had difficulty staying on task, doubted their own abilities in a one-on-one testing environment, and gave up easily. They sought to terminate the assessment, reacted inappropriately to failure, and required constant praise and encouragement to continue. Ongoing follow-up of this cohort into *References 43, 44, 46, 50-52, 66, 67, 80, 81, 90, 93, and 95.

372

WHITFIELD & GRUNAU

adolescence has highlighted difficulties with school, social interactions, self-esteem, and anxiety and depression. Another study examined the role of biologic factors (birth weight and perinatal and medical complications) and psychosocial factors (maternal attitudes, paternal involvement, mother’s personal state, marital adjustment, family relations, and socioeconomic status) in predicting long-term outcome of 90 Israeli adolescents born prematurely less than 1500-g birth weight.%,55 The low-birth weight group were compared with 90 full-birth weight adolescent controls. The low-birth weight group scored lower on all measures (IQ visual memory, hyperactivity, depression, generalized anxiety, and self-concept) except reading comprehension. In a regression analysis in this study, low birth weight accounted for 20% of the variance in the children’s ratings. There is a possibility that the premature infant’s multiple pain exposure might be a contributory factor to these findings; if so, this is a cause of current clinical concern.

ANIMAL CORRELATES

Exposure to repeated episodes of pain in neonatal rat pups can cause long-term behavioral changes that can be evaluated when they grow to adulthood. The effects of pain exposure in neonatal animals can also be compared with similar pain exposure inflicted when the central nervous system has developed in the adult. There is now an extensive literature regarding similarities in development of the pain system in rat pups and human newborns.24,82 In one study, pronounced cutaneous skin hyperinnervation was observed following skin wounds in neonatal rats, which was much greater and more prolonged than that occurring after skin wounds in older or adult rats.” In another study neonatal rat pups stimulated with needle prick stimuli four times a day from birth to 7 days old were noted to have significantly lower pain thresholds at 16 and 22 days of age compared with rat pups receiving a similar l4 Neonatally pain stimulated rats regimen of tactile stimulation only.12* in adulthood had an increased preference for alcohol, increased anxiety, defensive withdrawal behavior, a prolonged memory for chemosensory cues, no differences in corticosterone or adrenocorticotrophic hormone responses to an emotional stressor, and increased expression of Fos in the somatosensory cortex following exposure to a hot plate. In the rat pup, repetitive neonatal pain may cause decreased pain threshold during development, leading to stress vulnerability and anxiety-mediated adult behavior correlated with early onset cognitive deficits. Changes in brain development may lead to such behavioral differences very similar to the cognitive and behavioral characteristics described in long-term followup studies of survivors of prolonged neonatal intensive care.

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

373

ATTENTION, AROUSAL, AND SELF-REGULATION

Arousal and regulation of attention in preterm infants suggest al87 Orienting and distress in intered neuropsychological fun~tioning.~~, fancy may be closely linked systems.75Preterm infants show differences in their range and threshold for optimal states of arousal, which facilitate selective and sustained attention and information encodingz9 In social interactions, very low-birth weight infants have been described as more difficult to bring to an attentive state? giving mixed cues of attention and distress during en face mother-infant facial exchanges,59with those of higher biologic risk even more apt to switch from attention to distress than equally premature infants of lower biologic risk,21,22 and unavailable The literature on early self-regulation to stimulation and easily suggests that chronically stressful situations may inhibit development of integrative self-regulatory processes.79,96 Early sustained pain may be one factor disrupting development of infant self-regulatory patterns. Developmentally, system disregulation in early infancy may be precursors of later dysfunction in attention and higher-order executive processes. Preterm infants show delays and deficits in selective attention in infancy and are less likely to orient to or spend time exploring novel stimuli,29,73, 76, 77 habituate less efficiently to visual stimuli,17 and encode information less efficiently,% when compared with term infants at the same postconceptional age. Comparable difficulties in attention and information processing, however, are evident across childhood. Differences between ELBW and socially comparable term-born children in attention and response to unfamiliar or novel tasks are pervasive across childhood.37,38, 95 At age 3 years corrected chronologic age and 4.5 and 8.5 years chronologic age, during cognitive assessment independent of ability level, ELBW children preferred easy tasks, distrusted their own ability, reacted to failure unrealistically, and needed constant praise and encouragement. Poor regulation of arousal in infancy may reflect alteration of multiple systems that change in expression from infancy through school age. Individual differences in response to mild stressors are reflected in physiologic activity68and arousal of the hypothalamic-pituitary-adrenocortical system." In experimental animal studies, the hypothalamic-pituitary-adrenocortical axis can be permanently altered through environmental manipulations in the.neonata1 period.94In human infants at 32weeks postconceptional age baseline resting heart rate was related to prior exposure to invasive procedural pain, independent of gestational age at birth and illness severity (Grunau et al, unpublished data). This suggests that basal arousal in ELBW infants may be altered, at least in the neonatal period, by early prolonged exposure to neonatal pain. Because of the potential links between development of attention and arousal, this may be one mechanism of long-term behavioral alterations in former ELBW children. Differences in child rearing, parental sensitivity to child cues, and other environmental factors, however, are expected to ameliorate any such long-term effects of early pain.

374

WHITFIELD & GRUNAU

SOME ASPECTS OF THE PAIN MANAGEMENT DILEMMA

The previous information provides some compelling reasons to be concerned about the management of procedural pain in a NICU setting. It is beyond the scope of this article to explore critically the evidence behind the current clinical literature regarding the use of analgesics and sedation in ELBW babies undergoing neonatal intensive care. Prevention of pain and suffering have to be balanced against the risks of pharmacotherapy and other techniques designed to reduce pain and stress for these very fragile and evolving patients. It is perhaps individual unit interpretations of these risks and benefits and the paucity of information on which to form an evidence-based policy that explains the wide degree of variation in current practice from one NICU to another. Pharmacologic choices usually are between either pain-specific analgesia (e.g., morphine or other opiates) or sedation (e.g., midazolam or other benzodiazepine) to treat responses to noxious stimuli. Midazolam, which is currently used widely in neonatal intensive care, is a sedative and does not act specifically as an analgesic, whereas opiates are potentially addictive central analgesics that may not affect some of the local neurogenic effects of pain described previously. Both varieties of pharmacologic agents might permanently modify the course of the developing central nervous system, just as there appears to be evidence for pain exposure having this type of effect. The situation is further complicated by recent evidence from animal research that morphine may act differently in the brain in the presence of pain than it does in the absence of pain.69Administration of morphine only when pain can be demonstrated may be critical for preventing unwanted long-term side effects of opioid use. Conversely, administering midazolam when pain is present is inappropriate, because the detrimental physiologic spin-off effects of pain are not likely to be contr~lled.'~ SUMMARY

This article explores the literature concerning responses to pain of both premature and term-born newborn infants, the evidence for shortterm and long-term effects of pain, and behavioral sequelae in individuals who have experienced repeated early pain in neonatal life as they mature. There is no doubt that pain causes stress in babies and this in turn may adversely affect long-term neurodevelopmental outcome. Although there are methods for assessing dimensions of acute reactivity to pain in an experimental setting, there are no very good measures available at the present time that can be used clinically. In the clinical setting repeated or chronic pain is more likely the norm rather than infrequent discrete noxious stimuli of the sort that can be readily studied. The wind-up phenomenon suggests that, exposed to a cascade of procedures as happens with clustering of care in the clinical setting in an

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

375

attempt to provide periods of rest for stressed babies, an infant may in fact perceive procedures that are not normally viewed as noxious, as pain. Pain exposure during lifesaving intensive medical care of ELBW neonates may also affect subsequent reactivity to pain in the neonatal period, but behavioral differences are probably not likely to be clinically significant in the long term. Prolonged and repeated untreated pain in the newborn period, however, may produce a relatively permanent shift in basal autonomic arousal related to prior NICU pain experience, which may have long-term sequelae. In the long run, the most significant clinical effects of early pain exposure may be on neurodevelopment, contributing to later attention, learning, and behavior problems in these vulnerable children. Although there is considerable evidence to support a variety of adverse effects of early pain, there is less information about the long-term effects of opiates and benzodiazepines on the developing central nervous system. Current evidence reviewed suggests that judicious use of morphine for adjustment to mechanical ventilation may ameliorate the altered autonomic response. It may be very important, however, to distinguish stress from pain. Animal evidence suggests that the neonatal brain is affected differently when exposed to morphine administered in the absence of pain than in the presence of pain. Pain control may be important for many reasons but overuse of morphine or benzodiazepines may have undesirable long-term effects. This is a rapidly evolving area of knowledge of clear relevance to clinical management likely to affect long-term outcomes of high-risk children. References 1. Abu-Saad HH, Bours GJ, Stevens B, et al: Assessment of pain in the neonate [review].

Semin Perinatol 22:402, 1998 2. Als H Neurobehavioral organization of the newborn: Opportunity for assessment and intervention [review]. NIDA Res Monogr 114:106, 1991 3. Ambuel 8, Hamlett KW, M a n CM, et al: Assessing distress in pediatric intensive care environments: The COMFORT scale. J Pediatr Psycho1 1795, 1992 4. Anand KJ: Neonatal stress responses to anesthesia and surgery [review]. Clin Perinatol 17207, 1990 5. Anand KJ, Aynsley-Green A. Measuring the severity of surgical stress in newborn infants. J Pediatr Surg 23:297, 1988 6. Anand KJ, Carr DB: The neuroanatomy, neurophysiology, and neurochemistry of pain, stress, and analgesia in newborns and children [review]. Pediatr Clin North Am 36795, 1989 7. Anand KJ, Hickey PR Pain and its effects in the human neonate and fetus [review]. N Engl J Med 3171321, 1987 8. Anand KJ, Sippell WG, Aynsley-Green A: Randomised trial of fentanyl anaesthesia in preterm babies undergoing surgery: Effects on the stress response. Lancet 1:62, 1987 9. Anand KJ, Sippell WG, Schofield NM, et al: Does halothane anaesthesia decrease the metabolic and endocrine stress responses of newborn infants undergoing operation? BMJ 296:668, 1988 10. Anand KJS: Analgesia and sedation in ventilated neonates. Neonatal Respiratory Diseases 5:1, 1995 11. Anand KJS Clinical importance of pain and stress in preterm neonates. Biol Neonate 73:1, 1998

376

WHITFIELD & GRUNAU

12. Anand KJS, Coskun SV, Thrivikraman KV, et al: Long term behavioral effects of repetitive pain in neonatal rat pups. Physiol Behav 66:627, 1999 13. Anand KJS, McIntosh N, Lagercrantz H, et al: Analgesia and sedation in preterm neonates who require ventilatory support. Arch Pediatr Adolesc Med 153:331, 1999 14. Anand KJS, Thrivikraman KV, Engelman M, et al: Adult rat behaviour and stress responses following pain in the neonatal period. Pediatr Res 3757A, 1995 15. Aynsley-Green A Editorial: Pain and stress in infancy and childhood-where to now? Paediatr Anaesth 6167, 1996 16. Barker DP, Rutter N: Exposure to invasive procedures in neonatal intensive care unit admissions. Arch Dis Child 72:F47, 1995 17. Bornstein MH: Habituation as a measure of visual information processing in human infants: Summary, systemization, and synthesis. In Gottlieb G, Krasnegor N (eds): Development of Audition and Vision During the First Year of Postnatal Life: A Methodological Overview. Norwood, NJ, Ablex, 1985, p 253 18. Craig KD, Grunau RVE, Aquan-Assee J: Judgement of pain in newborns: Facial activity and cry as determinants. Canadian Journal of Behavioural Science 20442, 1988 19. Craig KD, Hadjistavropoulos HD, Grunau RVE, et al: A comparison of two measures of facial activity during pain in the newborn child. J Pediatr Psychol 19305, 1994 20. Craig KD, Whitfield MF, Grunau RVE, et al: Pain in the preterm neonate: Behavioural and physiological indices. Pain 52:287, 1993 21. Eckerman CO, Oehler JM, Hannan TE, et al: The development prior to term age of very prematurely born newborns’ responsiveness in en-face exchanges. Infant Behavior and Development 18:283, 1995 22. Eckerman CO, Oehler JM, Medvin MB, et al: Premature newborns as social partners before term age. Infant Behavior and Development 1755, 1994 23. Engel GL: Psychogenic pain and the pain prone patient. Am J Med 262399, 1959 24. Fitzgerald M: Development of pain pathways and mechanisms. In Anand KJS, McGrath PJ (eds): Handbook of Experimental Pharmacology: The Pharmacology of Pain. Berlin, Springer Verlag, 1997, p 447 25. Fitzgerald M, Millard C, McIntosh N Cutaneous hypersensitivity following peripheral tissue damage in newborn infants and its reversal with topical anaesthesia. Pain 39:31, 1989 26. Fitzgerald M, Shaw A, MacIntosh N: Postnatal development of the cutaneous flexor reflex: Comparative study of preterm infants and newborn rat pups. Dev Med Child Neurol 30:520, 1988 27. Franck LS, Miaskowski C: Measurement of neonatal responses to painful stimuli: A research review. J Pain Symptom Manage 14343, 1997 28. Gagnon RE, Leung A, Macnab AJ: Variations in regional cerebral blood volume in neonates associated with nursery events. Am J Perinatol 16:7, 1999 29. Gardner JM, Karmel BZ: Attention and arousal in preterm and full-term neonates. In Field T, Sostek A (eds): Infants Born at Risk: Physiological, Perceptual and Cognitive Processes. New York, Grune & Stratton, 1983, p 69 30. Grunau RE: Do early pain experiences have long term consequences-evidence from the clinic? Presented at the Fourth International Symposium on Paediatric Pain, Helsinki, 1997 31. Grunau RE: Long term effects of pain in pain and pain management during infancy. Research and Clinicals Forums 20:19, 1998 32. Grunau RE, Holsti L, Whitfield M, et al: Are twitches, startles and other body movements pain indicators in extremely low birthweight infants? Clin J Pain, in press 33. Grunau RE, Oberlander T, Holsti L, et al: Bedside application of the Neonatal Facial Coding System in pain assessment of premature neonates. Pain 76:277, 1998 34. Grunau RE, Whitfield MF, Petrie J: Children’s judgements about pain at age 8-10 years: Do extremely low birthweight (< or = 1000 g) children differ from full birthweight peers? J Child Psychol Psychiatry 39:587, 1998 35. Grunau RVE, Craig KD: Facial activity as a measure of neonatal pain expression. In Tyler DC, Krane EJ (eds): Advances in Pain Research and Therapy, vol 15. New York, Raven Press, 1990, p 147

BEHAVIOR, PAIN PERCEPTION, AND THE ELBW SURVIVOR

377

36. Grunau W E , Craig KD: Pain expression in neonates: Facial action and cry. Pain 28:395, 1987 37. Grunau RVE, Kearney SM, Whitfield MF: Language development at three years in preterm children of birth weight 4 0 0 0 grams. British Journal of Disorders in Communication 25:173, 1990 38. Grunau RVE, Whitfield MF, Fryer LE, et al: Behavior at 3 years and 4 1 / 2 years in high risk preterm children and full term controls. Presented at the Northwest Society for Developmental and Behavioral Pediatrics, Seattle, 1990 39. Grunau RVE, Whitfield MF, Petrie JH, et al: Early pain experience, child temperament and family characteristics as precursors of somatization: A prospective study of preterm and fullterm children. Pain 56356, 1994 40. Grunau RVE, Whitfield MF, Petrie JH: Pain sensitivity and temperament in extremely low-birthweight premature toddlers and preterm and full term controls. Pain 58:341, 1994 41. Guinsburg R, Berenguel RC, de Cassia Xavier R, et al: Are behavioral scales suitable for preterm and term neonatal pain assessment? In Jensen TS, Turner JA, WiesenfeldHallin Z (eds): Proceedings of the 8th World Congress on Pain, Progress in Pain Research and Management, vol 8. Seattle, IASP Press, 1997, p 893 42. Gunnar MR, Brodersen L, Krueger K, et al: Dampening of adrenocortical responses during infancy: Normative changes and individual differences. Child Dev 67877, 1996 43. Hack M, Klein N, Taylor H G Long term developmental outcomes of low birthweight infants. Future Child 5:176, 1995 44. Hack M, Taylor G, Klein N, et al: School age outcomes in children with birthweights under 750 grams. N Engl J Med 331:753, 1994 45. Hadjistavropoulos H, Craig KD, Grunau RE, et al: Judging pain in infants: Behavioral, contextual, and developmental determinants. Pain 73:319, 1997 46. Hall A, McLeod A, Counsel1 C, et al: School attainment, cognitive ability and motor function in a total Scottish very-low birthweight population at eight years: A controlled study. Dev Med Child Neurol371037, 1995 47. Hughes MC, Zimin R Children with psychogenic abdominal pain in their families. Clin Pediatr 17569, 1978 48. Johnston CC, Stevens BJ: Experience in a neonatal intensive care unit affects pain response. Pediatrics 98:925, 1996 49. Johnston CC, Stevens B, Craig KD, et al: Developmental changes in pain expression in premature, full-term, two- and four-month-old infants. Pain 52:201, 1993 50. Kitchen W, Campbell N, Carse E, et al: Eight-year outcome in infants with birth weight of 500 to 999 grams: Continuing regional study of 1979 and 1980 births. J Pediatr 118:761, 1991 51. Klebanov PK, Brooks-Gum J, McCormick MC: Classroom behaviour of very low birth weight elementary school children. Pediatrics 94:700, 1994 52. Klebanov PK, Brooks-Gunn J, McCormick M C School achievement and failure in very low birth weight children. Pediatrics 15:248, 1994 53. Lawrence J, Alcock D, McGrath P, et al: The development of a tool to assess neonatal pain. Neonatal Network 12:59, 1993 54. Levy-Shiff R, Einat G, Har-Even D, et al: Emotional and behavioural adjustment in children born prematurely. J Clin Child Psychol 23:323, 1994 55. Levy-Shiff R, Einat G, Mogilner M, et al: Biological and environmental correlates of developmental outcome of prematurely born infants in early adolescence. J Pediatr Psychol 19:63, 1994 56. Lollar DJ, Smits SJ, Patterson DL: Assessment of pediatric pain: An empirical perspective. J Pediatr Psychol 22:267, 1982 57. Low JA, Froese AB, Smith JT: Hypotension and hypoxaemia in the preterm newborn during the four days following delivery identdy infants at risk of echosonographically demonstrable cerebral lesions. Clin Invest Med 15:60, 1992 58. Mayes LC, Bomstein M H Attention regulation in infants born at risk. In Burack JA, E m s JT (eds): Attention, Development and Psychopathology. New York, Guildford Press, 1997, p 97

378

WHITFIELD & GRUNAU

59. McGehee LJ, Eckerman C O The preterm infant as a social partner: Responsive but unreadable. Infant Behavior and Development 6:461, 1983 60. McGrath PA, Seifert CE, Speechley KN, et al: A new analogue scale for assessing children’s pain: An initial validation study. Pain 64:435, 1996 61. McIntosh N: Pain in the newborn: A possible new starting point. Eur J Paediatr 156173, 1997 62. Minuchin S: Families and Family Therapy. Cambridge, MA, Harvard University Press, 1974 63. Norris S, Campbell LA, Brenkert S Nursing procedures and alterations in transcutaneous oxygen tension in premature infants. Nurs Res 31:330, 1982 64. Oberlander T, Grunau RVE, Whitfield MF, et al: Biobehavioral pain responses in former extremely low birthweight infants at 4 months corrected age. Pediatrics 105:6, 2000 65. Perlman J, Thach B: Respiratory origin of fluctuations in arterial blood pressure in premature infants with respiratory distress syndrome. Pediatrics 81:399, 1988 66. Pharaoh POD, Stevenson CJ, Cooke RWI, et al: Clinical and subclinical deficits at 8 years in a geographically defined cohort of low birthweight infants. Arch Dis Child 70264, 1994 67. Pharaoh POD, Stevenson CJ, Cooke RWI, et a1 Prevalence of behaviour disorders in low birthweight infants. Arch Dis Child 70:271, 1994 68. Porges S W Vagal tone: A physiological marker of stress vulnerability. Pediatrics 90:498, 1992 69. Rahman W, Fitzgerald M, Aynsley-Green A, et al: The effects of neonatal exposure to inflammation and/or morphine on neuronal responses and morphine analgesia in adult rats. 111 Jensen TS, Turner JA, Wiesenfeld-Hallin 2 (eds): Progress in Pain Research and Management. Seattle, IASP Press, 1997, p 783 70. Ramaekers VT, Casaer P, Daniels H: Cerebral hyperperfusion following episodes of bradycardia in the preterm infant. Early Hum Dev 34:199, 1993 71. Ramenghi LA, Wood CM, Griffith GC, et al: Reduction of pain response in premature infants using intraoral sucrose. Arch Dis Child 74F126, 1996 72. Reynolds ML, Fitzgerald M. Long term sensory hyperinnervation following neonatal skin wounds. J Comp Neurol 358:487, 1995 73. Rose SA: Differential rates of information processing in full-term and preterm infants. Child Dev 543189, 1983 74. Rose SA, Schmidt K, Riese ML, et al: Effects of prematurity and early intervention on responsivity to tactile stimuli: A comparison of preterm and fullterm infants. Child Dev 51:416, 1980 75. Rothbart MK, Posner MI, Rosicky J: Orienting in normal and pathological development. Dev Psychopathol 6:635, 1994 76. Ruff HA: Attention and organization of behavior in high risk infants. J Dev Behav Pediatr 7298, 1986 77. Ruff HA: The measurement of attention in high risk infants. In Vietze PM, Vaughan HG (eds): Early Identification of Infants with Developmental Disabilities. New York, Grune & Stratton, 1988, p 282 78. Rushforth JA, Levene MI: Behavioral response to pain in healthy neonates. Arch Dis Child 70:F174, 1994 79. Ryan RM, Kuhl J, Deci EL: Nature and autonomy: An organizational view of social and neurobiological aspects of self-regulation in behavior and development. Dev Psychopathol 9:701, 1997 80. Saigal S, Rosenbaum P, Stoskopf B, et al: Comprehensive assessment of the health status of extremely low birth weight children at 8 years of age: Comparison with a reference group. J Pediatr 125:411, 1994 81. Saigal S, Szatmari P, Rosenbaum P, et al: Cognitive abilities and school performance of extremely low birth weight children and matched term control children at age 8 years: A regional study. J Pediatr 118:751, 1991 82. Schellinck HA, Anand KJS: Consequences of early pain experience: Lessons from rodent models of newborn pain. In McGrath PJ, Finley GA (eds): Chronic Recurrent Pain in Children and Adolescents. Seattle, IASP Press, 1999, p 40

BEHAVIOR, PAIN PERCEPTION, A N D THE ELBW SURVIVOR

379

83. Scott CS, Riggs KW, Ling E, et al: Morphine pharmacokinetics and pain assessment in premature newborns. J Pediatr 135:423, 1999 84. Sigman M: Individual differences in infant attention: Relation to birth status and intelligence at 5 years. In Field T, Sostek A (eds): Infants Born at Risk: Physiological, Perceptual and Cognitive Processes. New York, Grune & Stratton, 1983, p 271 85. Stevens BJ, Johnston CC, Horton L: Factors that influence the behavioral pain responses of premature infants. Pain 59:101, 1994 86. Stevens 8, Johnston C, Petryshen P, et al: Premature Infant Pain Profile: Development and Initial Validation. Clin J Pain 12:13, 1996 87. Szatmari P, Saigal S, Rosenbaum P, et al: Psychiatric disorders at five years among children with birthweights d o 0 0 g: A regional perspective. Dev Med Child Neurol 32:954, 1990 88. Taddio A, Goldbach M, Ipp M, et al: Effect of neonatal circumcision of pain during vaccination. Lancet 345:291, 1995 89. Taddio A, Katz J, Ilersich AL, et al: Effects of neonatal circumcision on pain response during subsequent routine vaccination. Lancet 349:599, 1997 90. Taylor H Q Hack M, Klien N, et al: Achievement in children with birth weights less than 750 grams with normal cognitive abilities: Evidence for specific learning disabilities. J Pediatr Psycho1 20:703, 1995 91. Tronick EZ, Scanlon KB, Scanlon JW: Protective apathy, a hypothesis about the behavior organization and its relation to clinical and physiologic status of the preterm infant during the newborn period. Clin Perinatol 17125, 1990 92. Truog R, Anand KJ: Management of pain in the postoperative neonate [review]. Clin Perinatol 16:61, 1989 93. Vohr B, Garcia Coll C, Flanagan P, et al: Effects of intraventricular hemorrhage and socioeconomic status on perceptual, cognitive, and neurologic status of low birth weight infants at 5 years of age. J Pediatr 121:280, 1992 94. Weinberg J: Recent studies on the effects of fetal alcohol exposure on the endocrine and immune systems. Alcohol Alcohol Supp 2:401, 1994 95. Wlutfield MF, Grunau RVE, Holsti L: Extreme prematurity (<800 grams) at school age: Multiple areas of hidden disability. Arch Dis Child 77F85, 1997 96. Wilson BJ, Gottman J M Attention-The shuttle between emotion and cognition: Risk, resiliency, and physiological bases. In Hetherington EM, Blechman EA (eds): Stress, Coping and Resiliency in Children and Families. Mahwah, NJ, Erlbaum, 1996 97. Zahr LK, Balain S Responses of premature infants to routine nursing interventions and noise in the NICU. Nurs Res 44:179, 1995 98. Zemikow B, Michel E, Kohlmann G, et al: Cerebral autoregulation of preterm neonates: A non-linear control system. Arch Dis Child 70:F166, 1994

Address reprint requests to Michael F. Whitfield, MD Department of Paediatrics University of British Columbia B.C.’s Children’s Hospital, Room lRlO 4480 Oak Street Vancouver, BC V6H 3V4 Canada e-mail: [email protected]