Access the full text.
Sign up today, get DeepDyve free for 14 days.
BioscienceHorizons Volume 7 2014 10.1093/biohorizons/hzu006 Review ‘Neonates do not feel pain’: a critical review of the evidence Amy Marchant* Faculty of Biological Sciences, University of Leeds, UK *Corresponding author: Email: firstname.lastname@example.org Supervisor: Dr Anne King, Faculty of Biological Sciences; University of Leeds, UK. Up until 1985, the nervous system of the neonate was widely considered to be underdeveloped for pain sensation. Analgesia to alleviate distress from ‘painful’ procedures in which neonates were, and still are, subject to was often considered trivial. Pain in the neonate and (in some cases) the disabled neonate is especially hard to investigate, as they are unable to verbally com- municate. There is also no known direct biological marker of pain, only behavioural and stress-related physiological corre- lates. This critical review gives evidence for and against the hypothesis ‘Neonates do not feel Pain’ . Evidence of both the neonatal response to analgesics and long-term effects of neonatal pain are also investigated, with the aim of further supporting or falsifying the hypothesis. Convergence of the observations covered in this review show that most, if not all, studies are in favour of pain-related behaviour and physiology in the neonate, both of which having a similar phenotype to that seen in the older infant and adult. The evidence investigated in this review also supports the hypothesis that cortical development appears to accommodate the subjectivity of pain, but it is not vital for pain experience. Further data and theory have the potential to bring more invaluable evidence to the table regarding whether or not the neonate is able to feel pain. Key words: neonate, pain, behaviour, physiology, cognition, analgesia Received 5 December 2013; accepted 13 June 2014 Introduction (Cote, Lerman and Todres, 2009). This dry spell in neonatal pain research was alleviated in 1980, with a study by Anand Hypothesis: ‘Neonates do not feel pain’ et al. (1985) leading to the observation of pain-induced responses and death due to endocrine shock during neonatal During the vulnerable neonatal period from birth onwards in surgery, emphasizing the importance of anaesthesia. which forced immobilization, feeding tube insertion, heel Convergence in human and animal study data has led to lance and other such invasive procedures often take place recent advances in the field of neonatal pain ( Johnston and (Carbajal et al., 2008), one would assume that pain sensitiv- Walker, 2003). ity is greatest. Up until the late 1980s, however, it was widely considered among clinicians that neonatal pain experience Despite many findings pointing towards some kind of was non-existent due to the belief that they were undevel- unpleasant sensation being felt in the neonate on administra- oped and incapable (Owens and Todt, 1984). The 20th cen- tion of noxious stimuli, conflicting experimental results tury appeared to be a time in which clinicians were more (potentially due to the subjective nature of whether or not we concerned about foetal distress. This is reflected by the sig - think a subject is in pain, for example) still prevail in the nificant lack of studies from 1920 until 1980 investigating discussion of neonatal pain. The significant limitations of pain in the neonate; curare often being used as a surgical pain research such as the inability of neonates (and often the muscular paralysis agent up until 1985, making pain-related handicapped) to verbally vocalize whether or not they are in protest or movement impossible regardless of its existence pain can often lead to contrasting ideas. In addition, only © The Author 2014. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Research article Bioscience Horizons • Volume 7 2014 non-invasive, medically necessary methods of pain induction this view and support the hypothesis that the cortex is not are permitted in humans (as per the ethical restrictions laid necessary for pain, therefore undermining the hypothesis that down by the International Association for the Study of Pain neonates do not feel pain. Facial action coding and auto- (2013)). Direct quantitative measurement of pain is also not nomic responses in reaction to noxious heel lance in those possible due to the lack of pain ‘centre’ and biological <32 weeks were found to be subcortically mediated, due to marker; therefore, objective indicators such as self-report (in the lack of difference in response between severely brain- adults) and facial expressions (in neonates such as brow injured and control infants. It is important to bear in mind, bulge, crease and furrow) are heavily relied upon (Grunau however, that no two lesioned brains are the same, and differ- and Craig, 1987). Some physiological correlates, however, ing connections within the ‘pain matrix’ may have been can often be indicative of the experience, such as respiration, affected in these studies, questioning data validity. heart rate and serum cortisol levels (Hummel and van Djik, Evidence from Fabrizi et al. (2011), however, further falsi- 2006). Promising acoustic studies such as that by Branco et al. fies the hypothesis that neonates do not feel pain through the (2007) have been questioned, with some believing outward use of electroencephalography (EEG) to document the change communication and inner perception are not correlated. from non-specific, consistently distributed bursts to modality- Modern breakthroughs such as non-invasive neuroimaging specific, localized potentials across the human neonatal cor - (specifically functional magnetic resonance imaging or fMRI) tex. This was determined by analysing responses to touch have brought invaluable new pain-related evidence to the and heel lance in comparison to gestational age (GA), the table; this technology is vital for our further understanding. range of which being from 28 to 45 weeks. Many subjects Recent evidence surrounding the following points will be were used (n = 48) and chosen using strict eligibility criteria critically analysed in this review, with the hypothesis, (increasing statistical power and result validity), with burst ‘Neonates do not feel pain’, in mind. characterization being identified in accordance to their char - acteristic frequency, polarity, amplitude and duration. Sixty At what point is the switch in sensory EEG recordings were analysed in total, the results of which processing from tactile to noxious were compared with that previously recorded in the neonatal sensation? rat cortex (Diesch et al., 2009). The critical period was from 35 to 36 weeks for nociceptive and 36 to 37 weeks for tactile It is both a valid and common view that neonatal pain is only processing, and the values for nociceptive and tactile process- viable due to functional connections between brain structures ing were not significantly different (analysis of variance obligatory for perception and the periphery. Neurons of the (ANOVA): P = 0.96). This suggests that specific touch–pain spinothalamic projection begin to differentiate on embryonic discriminatory circuits are in place from 35 to 37 weeks of day 12 (E12) in rats (Altman and Bayer, 1984), with central gestation. This finding is opposed to the view that pain is felt projections of both myelinated and unmyelinated afferents only when free nerve endings and spinal projections develop fully entering the dorsal horn by E19 (Jackman and Fitzgerald, at 23–25 weeks (Fitzgerald, 1987) (Fig. 1). 2000), having adult-like function and morphology at P2 (Woodbury et al., 2008). Both the spinothalamic tract and Fabrizi et al. (2011) propose that the transition in process- free nerve endings are present at 8 weeks of gestation ing from tactile to noxious may be related to thalamocortical (Fitzgerald, 1987); these being regarded by some as the mini- connection development and the formation of callosal mum necessary framework to support pain processing. This pathways. A key role of pacemaker neurons within the view is strengthened by reports of whole body movement maturing lamina I of the spinal cord was also implied; these away from a stimulus due to an appearance of perioral inner- being thought to provide an endogenous drive to developing vation at this point in time (Myers et al., 2004); however, this pain circuitry (Fabrizi et al., 2011). This transition could hypersensitive tactile response may be compensating for the reasonably be considered as the first stage in the development immaturity of the nociceptive system and not necessarily con- of central pain processing. Interestingly, postnatal age at firming the experience of pain. This view of tactile hypersen - birth was not considered as a variable in this analysis—a sitivity over pain sensation is reinforced by the lack of factor that has been proposed to affect neonatal pain (Vinall thalamocortical projections and laminar structure at this et al. 2012). stage, the first sign of which appears at 12–16 weeks of gesta - tion (Fitzgerald, 1987). The foetal hormonal stress response The importance of spontaneous and evoked neuronal can be observed at 18 weeks (Giannakoulopoulos et al., bursting in functional circuit formation was further empha- 1994), and from 20 weeks, both the cortex and connections sized by Slater et al. (2010); these bursts being thought to between the thalamus and cortex are present (Garel et al., precede the onset of sensory function, emphasizing the criti- 2001). By 23–25 weeks, free nerve endings and spinal cord cal role of the cortex in relation to pain function. In contrast projections are fully mature (Fitzgerald, 1987); therefore, to this, hydranencephaly-inflicted children aged 4–21 months, pain sensation and perception are theoretically possible and who lacked significant regions of cortex due to their condi - question the view that pain involves sensory, emotional and tion, have also displayed pain-related behaviour (Jones and cognitive factors that are only fully mature after birth (Loeser France, 1978). Nandi et al. (2002) showed that ablation of and Melzack, 1999). Oberlander et al. (2002) have reinforced the thalamus affects pain perception. Taken together, these 2 Bioscience Horizons • Volume 7 2014 Research article Figure 1. The relationship between response type and non-specific neuronal burst/modality-specific potentials evoked by tactile and noxious stimuli of the heel, in relation to GA and topographical distribution. Touch and noxious heel lance are displayed as more likely to evoke a tactile and nociceptive-specific potential than a neuronal burst ~35–37 weeks of gestation. (Reprinted from Fabrizi et al. (2011), Copyright (2011), with permission from Elsevier). findings challenge cortico-centric views of pain function and terms (Merskey and Bogduk, 1994). Most individuals suggest a subcortical developmental route for the appearance associate the word ‘pain’ with early life injury, as suggested of pain function. by Craig and Korol, 2008, implying that memory has a role. The beginning of declarative memory use (due to prefrontal According to the International Association for the Study cortex and hippocampal development) at around 8–16 weeks of Pain, pain is an unpleasant sensory and emotional experi- of postnatal age (Richmond and Nelson 2007) allows the ence associated with tissue damage, or described in such storing of learned information. Evidence supporting the fact 3 Research article Bioscience Horizons • Volume 7 2014 that early pain sensation is mediated subcortically is stronger To conclude, physiological, haemodynamic and behav- than evidence for declarative pain memory in the neonate ioural data both before analgesia and the alterations in pain- (Nandi et al. 2002), however. It is important to note that the related response after analgesia indicate that neonates react development role of cortex and its relationship with subcor- to noxious stimuli, experiencing stress and displaying stereo- tex in pain function (and consciousness) is hugely controver- typical pain-induced responses. More research is required (in sial. The consensus view is that 18 weeks is too early for the addition to that by Segato et al. (1997)) to understand the appearance of pain function in the healthy foetus, and it is apparent pain-alleviating mechanism of sucrose, with any thought to be roughly 24–26 weeks of gestation. adverse effects, such as that seen in paracetamol (Cuzzolin, Antoniccu and Fanos, 2013). It is reasonable to suggest that Neonatal pain management regardless of whether or not the above analgesic interven- tions, including opioids, are purely a distraction mechanism Effective neonatal pain management remains a controversial as opposed to a direct analgesic—widely observed stereotypi- topic, despite major advances in pharmacological, physiologi- cal pain-related responses similar to that seen in the adult are cal and behavioural neonatal pain interventions within the clearly alleviated in the neonate on their administration, thus last few decades. Is it reasonable to suggest that the idea of falsifying the hypothesis. In addition to this, the long-term neonatal pain existence is valid if the stereotypical pain- effects of analgesia do not directly contribute to our under- related responses are alleviated by analgesics? Opioids are standing as to whether or not neonates feel pain. The long- routinely used in the Neonatal Intensive Care Unit (NICU), term effects of morphine are difficult to evaluate due to the and their pharmacokinetics and pharmacodynamics in infants confounding factors of illness and prematurity on neurode- have been studied intensely. An early study by Quinn et al. velopmental outcome, with some subjects having received (1993) found morphine to evoke a significant decrease in more morphine than others as a neonate. The prevention of pain-associated plasma adrenaline in pre-term neonates; how- long-term hypersensitivity using pre-emptive morphine is ever, insignificant haemodynamic and behavioural differences unclear, with the results of the only study examining neuro- were found. It is important to note that the efficacy of the development after NICU pre-emptive administration in those behavioural scale used had not yet been validated in pre-term aged 5–6 years being inconclusive due to low sample size neonates, this potentially questioning the validity of these (MacGregor et al., 1998). behavioural findings. A multidimensional assessment by Guinsburg et al. (1998) used strict eligibility screening to obtain 22 mechanically ventilated pre-term infants aged Evidence for long-term effects ≤32 weeks with a postnatal age of 12–48 h. These infants of the neonatal pain experience were studied 30 and 60 min after a single administration of The role of neuronal activity and the importance of the the opioid, fentanyl (3 mg/kg) or placebo, measuring the neonatal period to sensory development have been highly physiological and behavioural pain and stress indicators asso- documented. Significant neuronal development takes place ciated with mechanical ventilation (i.e. non-acute pain). High postnatally, with both structural and functional alterations of basal cortisol, growth hormone and lactate were present sensory connections occurring (Walker, Tochiki and Fitzgerald, before treatment, with behavioural scales indicating pain. 2009). Pre-term neonates exhibit low tactile threshold—their After fentanyl, minimum and maximum heart rate decreased system physiology being unstable and potentially rendering (due to increased vagal tone as per most opioids (Garofalo them more vulnerable to the effects of repeated invasive et al. 2008)) and growth hormone level increased, with behav- procedures (Grunau, 2002). ioural postoperative comfort score increasing and neonatal facial coding system score decreasing. Despite the fact that the Multiple studies have observed pain-induced activation of behavioural findings were purely qualitative, (assessed thr - the hypothalamic-pituitary-adrenal axis, immune system and ough bedside observation, video playback and still images) autonomic nervous system in both the pre-term and the full- and that even premature neonatal handling has been shown to term neonate (Neveu et al., 1994; Morison et al., 2001). This induce pain, it is clear that these findings imply a decrease in often leads to a general increase in stress response mediators; non-acute pain-related responses after morphine administra- their effects often being maintained beyond that which is nor- tion. Further studies have confirmed these results and contrib - mal, switched off prematurely or ‘overused’ due to multiple uted to falsifying the hypothesis that neonates do not feel stressors. Determining the specific effects of neonatal pain is pain, with Alencar et al. (2012) also observing a decrease in challenging due to the multiple NICU stressors contributing pain-related response and insignificant differences in arterial to high ‘allostatic load’ (the cumulative stress upon the neo- blood gas levels after fentanyl administration. The above nate (McEwen, 1999)). Increased knowledge of the subjects’ studies by Guinsburg et al. (1998) and Alencar et al. (2012) previous medical and environmental conditions could lead to indicate that neonatal opioid administration may not only better analysis of the effects of neonatal pain. decrease suffering that may potentially lead to long-term adverse effects, but also stabilize blood pressure, divert energy Chu et al. (2012) analysed the adult rat following neonatal sources to growth and have a beneficial effect on the clinical noxious insult, finding alterations in pain sensation, an stability of critically ill pre-term neonates. increase in resting blood pressure and weakened cardiovascular 4 Bioscience Horizons • Volume 7 2014 Research article for by including data at night to depict changes in autonomic activity throughout the day, increasing the relevance of the findings. Chu et al. (2012) implanted the telemetry device in the abdominal aorta of the mice studied; however, the reli- ability of the cardiovascular data obtained from this site has been questioned (Kaidi et al., 2007), left carotid implantation being considered far superior and more appropriate for mice and rat studies. This information may benefit future studies similar to Chu et al. (2012) by increasing the reliability and validity of pain-related cardiovascular recordings. According to some authors, evidence of hypersensitivity and loss of car- diovascular control sometime after neonatal pain-related stress supports the existence of pain sensation in the neonate (Morison et al., 2001; Beggs et al., 2012). Further replication of this recent study is required to further support these find - ings and to ensure that these results fully correlate to the human neonate. Handling and early maternal separation can also result in visceral and somatic hyperalgesia as found by Coutinho et al. (2002); therefore, future studies similar to this should account for the environmental history of the rats to obtain more Figure 2. Effects of a neonatal carrageenan injection on left hindpaw reliable results. withdrawal threshold following intraplantar CFA. Six hours (CFA 6 h) Long-term structural effects of the neonatal pain experi- and 24 h after CFA injection are shown. Saline-treated rats = 9 and ence have also been found. Neurotrophins are vital for the carrageenan-treated rats = 9. The box and whisker plots are expressed as medians with first and third quartiles (boxes), and 10th and 90th development of sensory skin innervation, also controlling percentiles (vertical lines). *P < 0.05. (Reprinted from Chu et al. (2012), the survival and function of neurons both centrally and Copyright (2012), with permission from Elsevier). peripherally (Huang and Reichardt, 2001). NT-3 (of the Nerve Growth Factor family of Neurotrophins) has shown responsiveness to nociceptive stress in adulthood. At P1, to be critical for cutaneous sensory nerve sprouting neonatal rats were administered either intraplantar saline or (Airaksinen et al., 1996). Beggs et al. (2012) built on an 0.25% carrageenan (1 µ l/g). Baseline cardiovascular variables early study by Reynolds et al. (1997) and found, through and 24-h responsiveness to Complete Freud’s Adjuvant (CFA) neonatal skin wounding, that Neurotrophin-3 (NT-3) is injection in the freely moving adults were recorded. Through highly regulated in the skin. A skin wound was induced in the use of Von Frey Filaments, it was found that carrageenan- both rats and mice at P1, with regional cutaneous innerva- treated rats exhibited generalized basal hypoalgesia (in agree- tion being analysed at P7 and compared with naive animals. ment with Bruehl and Chung (2004)) and localized The wounding induced a 3-fold up-regulation of protein hyperalgesia following nociception induced by adult intra- gene product (PGP) 9.5 positive fibres in the skin and a plantar CFA (Fig. 2). Significantly higher basal blood pres - 25-fold increase in release of NT-3 in comparison to control. sure, baroreceptor sensitivity and parasympathetic activity The sensory neurite outgrowth induced by this NT-3 release after CFA injection were also found in these rats in compari- was significantly reduced through the use of specific antibod - son to controls. A potential theory behind these cardiovascu- ies to block NT-3 activity in dorsal root ganglion and skin lar findings is that neonatal nociceptive stress induces a co-cultures. This confirmed NT-3 as a requirement for sen - persistent hypertensive response, eventually causing the devel- sory neurite outgrowth, as did the absence of wound-induced oping brain to reset and increase its basal level. The measure- hyperinnervation in heterozygous transgenic mice created by ment of paw withdrawal threshold both 6 and 24 h after CFA the team (NT-3 + / − lacZ). It is thought that NT-3 stimulates injection was appropriate in comparison to previous similar sensory nerve sprouting through the binding of TrkC recep- studies. The telemetry technique used by Chu et al. (2012) tors on nerve terminals, the accompanying TrkA and TrkB allowed continuous cardiovascular recordings to be obtained receptor activation sustaining neuronal life and encouraging from conscious, unstressed and freely moving animals con- new sensory neurons to form. The low-affinity interactions tinuously. This method is significantly more valid than tail- of NT-3 with TrkA and TrkB require high NT3 concentra- cuff plethysmography, which requires physical restraint of the tions, explaining the dramatic decrease in innervation fol- animal that may itself affect blood pressure, a 70% reduction lowing neonatal skin wounding in heterozygous NT-3 in the number of animals required plus no need for any addi- mutant mice. Many precautions were taken by Beggs et al. tional data manipulation (Braga and Burmeister, 2011). The (2012) to improve validity. PGP 9.5 immunoreactivity was potential influence of circadian rhythm (an idea first investi - compared with age-matched littermate controls to obtain as gated by Zhang, Zannou and Sannajust (2000)) was accounted representative results as possible. Skin samples were also 5 Research article Bioscience Horizons • Volume 7 2014 taken immediately post-wounding to avoid other confound- Hohmeister et al. (2010) used fMRI to investigate changes ing effects, with unwounded tissue from the contralateral in cerebral processing in school-aged children (11–16 years) side of the same animal being used for comparison to who were born at either pre-term or full term (≤31 weeks GA increase data validity. Despite these measures, it is important or ≥37 weeks GA, respectively) and had neonatal nociceptive to bear in mind that neonatal injury models vary in duration, input within NICU. The findings were compared with that of extent and location of their injury. Newborn mice and rats, control children who were born at full term with no experi- such as those used in this experiment, are born at an earlier ence of NICU. This is the first neuroimaging study to depict stage of development in comparison to humans; therefore, an exaggerated neuronal response to pain in pre-term NICU this investigation is more likely to correspond with prema- children. During 30 s heat stimuli of individually adjusted ture infants than full-term infants. To conclude, neonatal moderate pain intensity, pre-term (but not term) NICU chil- wounding was found to increase NT-3 transcription, protein dren exhibited significantly higher brain activations (through levels and release 3–7 days post-injury in both rats and mice, greater number of voxel activation) in regions such as the cutaneous hypersensitivity being the result. This study primary somatosensory cortex, anterior cingulate cortex and focused not as much on pain sensation (in contrast to that by insula that were not significantly activated in controls; these Chu et al. (2012)), but on the structural and functional con- responses being pain specific as they were not observed dur - sequences of neonatal injury; this evidence being insufficient ing non-noxious thermal stimulation. This evidence indicates to support or deny the existence of neonatal pain. The results that pre-term noxious stimuli exposure induces greater acti- of this investigation could be considered valid due to the vation of sensory, affective and cognitive pain-related regions, many precautions taken; however, a longer time frame of the most robust alterations occurring in those implicated in investigation may have resulted in increasingly reliable evi- affective processing (Craggs et al., 2007). The continuous dence for long-term hyperinnervation into adulthood after pain ratings of the pre-term children revealed increased sensi- wounding. It is reasonable to suggest from this evidence that tization and a lack of habituation, implying neonatal pain prevention of local NT-3 up-regulation and therefore hyper- may persistently increase the gain within pain pathways. innervation may reduce the effects of neonatal tissue trauma Unfortunately, the fMRI session took place 2 years after in NICU or through surgery. initial psychophysical evaluation, and as no data on develop- mental alterations in pain sensitivity in NICU neonates were It is thought that anti-nociceptive structures may be under- available, normalization of pain threshold may have occurred. developed at birth, with a novel study by Brummelte et al. The various strengths of this study, however, appear to out- (2012) being the first to associate procedural NICU pain with weigh the limitations such as the use of extensive subject early neurodevelopment in those very pre-term. Greater pain exclusion criteria to increase validity plus the visual inspection was found to be associated with reduced white matter frac- of structural scans by the team to rule out severe morpho- tional anisotropy (therefore white matter integrity) and logical alterations. During stimulation, children continuously reduced subcortical grey matter N-acetylaspartate/choline rated stimulus intensity on a computerized visual analogue (increased grey matter neuronal loss), using 3D magnetic scale to track changes in perceived intensity and to direct resonance spectroscopic imaging paired with diffusion tensor attention continuously towards the stimuli, further increas- imaging (DTI). This neuronal loss may have been due to exci- ing validity. Overall, this study is an adequate evidence to totoxicity in overactive, immature neuronal networks. both disprove the hypothesis that neonates do not feel pain Previous DTI limitations found in other studies (the inability and confirm the existence of long-term effects. Further stud - of the model to cope with non-Gaussian diffusion (Assaf and ies should include larger sample sizes and fMRI sessions close Pasternak, 2008)) were abolished through combination with to the time in which psychophysical evaluation takes place. functional brain mapping in this study, resulting in reliable The influence of subtle structural brain abnormalities in the and non-invasive mapping of functional anatomy, reliably NICU pre-term cannot be ruled out, and the relationship indicating long-term effects of painful procedures in the neo- between structural abnormality and pain-related functional nate and indirectly falsifying the hypothesis that neonates do activation also requires further investigation not feel pain. A similar study by Vinall et al. (2012) using neuro- Conclusion imaging examined whether a greater number of neonatal skin- breaking events led to a decreased postnatal weight and The flurry of activity investigating the existence of neonatal head circumference (HC) early in life and at term-equivalent pain from 1985 onwards has led to the general consensus age in infants born very pre-term. It was found that greater that neonates do feel pain on the administration of noxious neonatal pain led to lower body weight (ANOVA, P = 0.01) stimuli such as heel lance and other procedures. The emer- and HC (ANOVA, P = 0.04) percentiles at 32 weeks GA, gence of fMRI has put a new slant on behavioural and physi- independent of other medical confounds. This study suggests ological studies, especially through the analysis of cortical that successful postnatal growth is dependent on the quantity responses to pain. It is clear to see that animal models have of painful procedures encountered during early NICU, show- proved essential for interpretation and confirmation of ing that brain maturation is affected by neonatal affliction, human findings, despite there being differences in complexity rather than degree of prematurity at birth. between species. 6 Bioscience Horizons • Volume 7 2014 Research article The investigation of analgesics has also proved to be very questioning the idea of neonatal pain may reflect the personal worthwhile in dissecting the existence of pain in the neonate, spiritual or religious views of the scientists involved. However, with pre- and post-analgesic responses further supporting its the more likely explanation may be due to ambiguous results functional existence, the fentanyl studies by Guinsburg et al. and techniques that make its existence slightly cloudier. Various (1998) and Alencar et al. (2012) being particularly poignant. noxious stimuli-induced inconsistencies emphasize the variabil- Could it be said that morphine and other opiods merely distract ity between individuals and the importance of repetitive single- the subject from the pain they are experiencing instead of subject analysis and sizeable study groups in pain research. directly targeting it? If this is found to be true, the use of anal- Neonates with neurological lesions have typically been excluded gesic efficacy to prove the existence of neonatal pain is ques - from both pain-related studies of the premature and full term, tionable, emphasizing the need for further research on this class possibly due to the vast inter-individual variation of lesions. Due of drugs and a further awareness of the clinical factors that to this, minimal knowledge of pain experience within these determine analgesic effectiveness. The importance of analgesic populations has been elucidated, this being a potential area for use is emphasized by structural studies such as that by Beggs further neonatal pain research. et al. (2012), with many more contributing significant evidence To conclude, given the strong evidence outlined above in of the long-term effects of pain experienced during the neonatal favour of stress and pain-related behaviour in the neonate, it period showing that long-term effects exist and are not to be is reasonable to believe that neonates do feel some form of taken lightly. The earlier an infant is subjected to pain, it seems, unpleasant sensation or pain-related stress on noxious stimuli the greater the potential for harm as emphasized by structural that have the phenotype of that which is felt in the infant and studies in the premature, such as that by Brummelte et al. adult. The evidence investigated in this review confirms that (2012). These effects have proved to be dependent on the cortical development appears to modulate and accommodate nature and timing of the insult, indicating plasticity at such a the subjectivity of pain, but it is not vital for pain experience. vulnerable time of life, the pain-related effects also occurring in adults—further falsifying the hypothesis that neonates do not feel pain. It is important to accept that with all premature and Author biography full-term studies, interpretation is often questionable as noci- A.M. finished her undergraduate degree in Neuroscience at ceptive events cannot be separated from other stresses such as the University of Leeds in July 2013. Her particular field of low tactile threshold, rapid brain development, GA and severe interest stems from lectures on the topic of pain sensation illness which may have either additive or synergistic effects on and the underlying mechanisms of sleep—finding the ways in neonatal developmental trajectories. Stress also plays a large which sleep and anaesthesia are linked, fascinating. Her part in our assumptions and is constantly linked to the pain future aspirations include doing an MSc in Neuroscience, response; however, while pain is stressful, it is questionable as studying postgraduate Medicine and being involved in bio- to whether stress is related to pain and whether some results of medical journalism at some point. A.M. wrote this paper and pain assessment are pain specific. has primary responsibility for the final content. Current evidence is pointing towards the existence of long- term effects, indirectly further proving the existence of neona- tal pain and hence falsifying my hypothesis. If there are indeed References cognitive, sensory and emotional elements to the overall pain Airaksinen, M., Koltzenburg, M., Lewin, G. et al. (1996) Specific subtypes experience, all of which are mandatory as hypothesized by of cutaneous Mechanoreceptors require neurotrophin-3 following Merskey (1986), pain can only be felt once in all parts of the peripheral target innervation, Neuron, 16, 287–295. hypothesized ‘pain matrix’ (Ploghaus et al., 1999) are formed, including the cortex. Convergence of data from previous stud- Alencar, A., Sanudo, A., Sampaio, V. et al. (2012) Efficacy of tramadol ver - ies implies this point is in fact after birth, however, how can sus fentanyl for postoperative analgesia in neonates, Archives of pain perception be purely learned if it is protective from harm- Disease in Childhood—Fetal and Neonatal Edition, 97, 24–29. ful stimuli? It is a reasonable assumption from previous stud- Altman, J. and Bayer, S. (1984) The development of the rat spinal cord, ies that pain-related responses can partly be mediated through Advances in Anatomy, Embryology and Cell Biology, 85, 1–66. reflex pathways within the brainstem and spinal cord and that pain experience is possible on establishment of thalamocorti- Anand, K., Brown, M., Causon, R. et al. (1985) Can the human neonate cal connections from gestational week 20 (Garel et al., 2001). mount an endocrine and metabolic response to surgery? Journal of All available evidence of cortical and analgesic response to Pediatric Surgery, 20 (1), 41–48. pain, in combination with evidence of the existence of long- term effects of pain, imply that pain sensation is functional Anand, K., Garg, S., Rovnaghi, C. et al. (2007) Ketamine reduces the cell approximately at this point, therefore falsifying the hypothe- death following inflammatory pain in newborn rat brain, Pediatric sis that neonates do not feel pain. Research, 62, 283–290. It is important to accept the fact that robust electrophysiolog- Assaf, Y. and Pasternak, O. (2008) Diffusion tensor imaging (DTI)-based ical evidence supporting my hypothesis will be hard to obtain, if white matter mapping in brain research: a review, Journal of not impossible. It is reasonable to suggest that investigations Molecular Neuroscience, 34 (1), 51–61. 7 Research article Bioscience Horizons • Volume 7 2014 Beggs, S., Alvares, D., Moss, A. et al. (2012) A role for NT-3 in the hyperin- Fitzgerald, M. (1987) The prenatal growth of fine diameter afferents into nervation of neonatally wounded skin, Pain, 153 (10), 2133–2139. the rat spinal cord—a transganglionic study, Journal of Comparative Neurology, 261, 98–104. Braga, V. and Burmeister, M. (2011) Applications of telemetry in small laboratory animals for studying cardiovascular diseases, modern Garel, C., Chantrel, E., Brisse, H. et al. (2001) Fetal cerebral cortex: normal telemetry. InTech, accessed at: http://www.intechopen.com/books/ gestational landmarks identified using prenatal MR imaging. modern-telemetry/applications-of-telemetry-in-small-laboratory- American Journal of Neuroradiology, 22, 184–189. animals-for-studying-cardiovascular-diseases (20 March 2013). Garofalo, N., Teixeira-Neto, J., Schwartz, D. et al. (2008) Effects of the opi - Branco, A., Fekete, S., Rugolo, L. et al. (2007) The newborn pain cry: oid remifentanil on the arrhythmogenicity of epinephrine in halo- descriptive acoustic spectrographic analysis, International Journal of thane-anesthetized dogs, Canadian Journal of Veterinary Research, Pediatric Otorhinolaryngology, 71, 539–546. 72 (4), 362–366. Bruehl, S. and Chung, O. (2004) Interactions between the cardiovascular Giannakoulopoulos, X., Sepulveda, W., Kourtis, P. et al. (1994) Fetal and pain regulatory systems: an updated review of mechanisms and plasma cortisol and β-endorphin response to intrauterine needling, possible alterations in chronic pain, Neuroscience and Biobehavioral Lancet, 344, 77–81. Reviews, 28, 395–414. Grunau, R. (2002) Early pain in preterm infants: a model of long-term Brummelte, S., Grunau, R., Chau, V. et al. (2012) Procedural pain and effects, Clinical Perinatology, 29 (3), 373–394. brain development in premature newborns, Annual Neurology, 71 Grunau, R. and Craig, K. (1987) Pain expression in neonates: facial action (3), 385–396. and cry, Pain, 28 (3), 395–410. Carbajal, R., Rousset, A., Danan, C. et al. (2008) Epidemiology and treat- Guinsburg, R., Kopelman, B., Anand, K. et al. (1998) Physiological, ment of painful procedures in neonates in intensive care units, The hormonal, and behavioral responses to a single fentanyl dose in Journal of the American Medical Association, 300 (1), 60–70. intubated and ventilated preterm neonates, Journal of Pediatrics, Chu, C., Yang, C., Lin, H. et al. (2012) Neonatal nociception elevated 132, 954. baseline blood pressure and attenuated cardiovascular responsive- Hohmeister, J., Kroll, A., Wollgarten-Hadamek, I. et al. (2010) Cerebral ness to noxious stress in adult rats, International Journal of processing of pain in school-aged children with neonatal nocicep- Developmental Neuroscience, 30 (6), 421–426. tive input: an exploratory fMRI study, Pain, 150 (2), 257–267. Cote, C., Lerman, J. and Todres, D. (2009) A Practice of Anaesthesia for Huang, E. and Reichardt, L. (2001) Neurotrophins: roles in neuronal devel- th Infants and Children, 4 edn. Saunders Elsevier, Philadelphia, PA. opment and function. Annual Review of Neuroscience, 24, 677–736. Coutinho, S., Plotsky, P., Sablad, M. et al. (2002) Neonatal maternal sepa- Hummel, P. and van Dijk, M. (2006) Pain assessment: current status and ration alters stress-induced responses to viscerosomatic nociceptive challenges, Seminars in Fetal Neonatal Medicine, 11, 237. stimuli in rat. American Journal of Gastrointestinal and Liver Physiology, 282 (2), G307–G316. International Association for the Study of Pain. (2013). Ethical guidelines for pain research in humans, accessed at: http://www.iasp-pain.org/ Craggs, J., Price, D., Verne, G. et al. (2007) Functional brain interactions AM/Template.cfm?Section = Research_in_Humans (3 April 2013). that serve cognitive-affective processing during pain and placebo analgesia. Neuroimage, 38 (4), 720–729. Jackman, A. and Fitzgerald, M. (2000) The development of peripheral hindlimb and central spinal cord innervation by subpopulations of Craig, K. and Korol, C. (2008) Developmental issues in understanding, dorsal root ganglion cells in the embryonic rat, Journal of assessing, and managing pediatric pain, In: G. A. Walco and K. R. Comparative Neurology, 418, 281–298. Goldschneider (eds) Pain in Children: A Practical Guide For Primary Care, Humana Press, New Jersey. Johnston, C. and Walker, C. (2003) The effects of exposure to repeat minor pain during the neonatal period on formalin pain behaviour Cuzzolin, L., Antoniccu, R. and Fanos, V. (2013) Paracetamol (acetamino- and thermal withdrawal latencies, Pain Research and Management, 8 phen) efficacy and safety in the newborn, Current Drug Metabolism, (4), 213–217. 14 (2), 178–185. Jones, R. and France, T. (1978) Electrophysiological responses in hydran- Diesch, T., Mellor, D., Johnson, C. et al. (2009) Electroencephalographic encephaly. American Journal of Ophthalmology, 85, 478–484. responses to tail clamping in anaesthetized rat pups, Laboratory Animals, 43, 224–231. Kaidi, S., Brutel, F., Van Deun, F. et al. (2007) Comparison of two methods (left carotid artery and abdominal aorta) for surgical implantation of Emeterio, E., Tramullas, M. and Hurle, M. (2006) Modulation of apoptosis radiotelemetry devices in CD-1 mice, Laboratory Animals, 41 (3), in the mouse brain after morphine treatments and morphine with- 388–402. drawal, Journal of Neuroscience Research, 83 (7), 1352–1361. Loeser, J. and Melzack, R. (1999) Pain: an overview, Lancet, 353, 1607–1609. Fabrizi, L., Slater, R., Worley, A. et al. (2011) A shift in sensory processing that enables the developing human brain to discriminate touch MacGregor, R., Evans, D., Sugden, D. et al. (1998) Outcome at 5–6 years of from pain, Current Biology, 21 (18), 1552–1558. prematurely born children who receive morphine as neonates, 8 Bioscience Horizons • Volume 7 2014 Research article Archives of Disease in Childhood—Fetal and Neonatal Edition, 79, Reynolds, M., Alvares, D., Middleton, J. et al. (1997) Neonatally wounded F40–F43. skin induces NGF-independent sensory neurite outgrowth in vitro, Developmental Brain Research, 102 (2), 275–283. McEwen, B. (1999) Allostasis and allostatic load: implications for neuro- psychopharmacology, Neuropsychopharmacology, 22, 108–124. Richmond, J. and Nelson, C. (2007) Accounting for change in declarative memory: a cognitive neuroscience perspective, Development Merskey, H. (1986) Pain terms, Pain, 3, S215–S221. Research, 27 (3), 349–373. nd Merskey, H. and Bogduk, N. (1994) Classification of Chronic Pain, 2 Segato, F., Castro-Souza, C., Segato, E. et al. (1997) Sucrose ingestion edn, IASP Press, Seattle, WA. pp. 209–14. causes opiod analgesia, Brazilian Journal of Medical and Biological Morison, S., Grunau, R., Oberlander, T. et al. (2001) Relations between Research, 30 (8), 981–984. behavioral and cardiac autonomic reactivity to acute pain in pre- Slater, R., Worley, A., Fabrizi, L. et al. (2010) Evoked potentials generated term neonates, Clinical Journal of Pain, 17 (4) 350–358. by noxious stimulation in the human infant brain, European Journal Myers, L., Bulich, L. A., Hess, P. et al. (2004) Fetal endoscopic surgery: of Pain, 14 (3), 321–326. indications and anaesthetic management, Best Practice & Research Stevens, B., Yamada, J., Lee, G. et al. (2013) Sucrose for analgesia in new- Clinical Anaesthesiology, 18 (2), 231–258. born infants undergoing painful procedures, Cochrane Database Nandi, D., Liu, X., Joint, C. et al. (2002) Thalamic field potentials during Systematic Review, 2013 (1), CD001069. deep brain stimulation of periventricular gray in chronic pain, Pain, Vinall, J., Miller, S., Chau, V. et al. (2012) Neonatal pain in relation to 97, 47–51. postnatal growth in infants born very preterm, Pain, 153 (7), Neveu, P., Deleplanque, B., Puglisi-Allegra, S. et al. (1994) Influence of 1374–1381. early life events on immune reactivity in adult mice, Developmental Walker, S., Tochiki, K. and Fitzgerald, M. (2009) Hindpaw incision in early Psychobiology, 27, 205–213. life increases the hyperalgesic response to repeat surgical injury: Oberlander, T., Grunau, R., Fitzgerald, C. et al. (2002) Does parenchymal brain critical period and dependence on initial afferent activity, Pain, 147 injury affect biobehavioral pain responses in very low birth weight (1–3), 99–106. infants at 32 weeks’ postconceptional age? Pediatrics, 110 (3), 570–576. Woodbury, C., Kullmann, F., McIlwrath, S. et al. (2008) Identity of myelin- Owens, M. and Todt, E. (1984) Pain in infancy: neonatal reaction to a heel ated cutaneous sensory neurons projecting to nocireceptive lami- lance, Pain, 20 (1), 77–86. nae following nerve injury in adult mice, Journal of Comparative Neurology, 508, 500–509. Ploghaus, A., Tracey, I., Gati, J. et al. (1999) Dissociating pain from its anticipation in the human brain, Science, 284, 1979–1981. Zhang, B., Zannou, E. and Sannajust, F. (2000) Effects of photoperiod Quinn, M., Wild, J., Dean, H. et al. (1993) Randomised double-blind con- reduction on rat circadian rhythms of blood pressure, heart rate and trolled trial of effect of morphine on catecholamine concentrations locomotor activity, American Journal of Regulatory, Integrative and in ventilated pre-term babies, Lancet, 342, 324–327. Comparative Physiology, 279 (1), R169–R178.
Bioscience Horizons – Oxford University Press
Published: Sep 25, 2014
Keywords: neonate pain behaviour physiology cognition analgesia
Access the full text.
Sign up today, get DeepDyve free for 14 days.