Introduction

Remodelling of nociceptive pathways after exposure to anoxious insult is hypothesised to contribute to the develop-ment of chronic pain that persists long after the insult hasbeen removed and healing has occurred [1, 2]. Experimentalmodels for both arthritic and neuropathic pain support theimportance of neural plasticity for persistent pain com-plaints [3, 4]. These data suggest that initial pain exposuresets the stage for subsequent increased pain sensitivity,which may be experienced as increased risk to a variety ofchronic pain conditions.

Until recently, newborns and children were consideredto be relatively insensate to painful stimuli. Furthermore,

a belief that babies would not remember their painfulexperience reduced efforts to minimise painful proceduresand use pre-emptive analgesia before procedures. Over thelast decade, studies have shown that children, includingnewborns, do perceive noxious stimuli as painful. By mea-suring heart rate, oxygen saturation, mean arterial pressureand behavioural response, Porter et al. [5] showed thatnewborns respond to painful stimuli, similar to adults. Inaddition, even very premature infants were shown to neg-atively respond to pain and to be able to differentiate stim-ulus intensity, with a stronger response shown for moreinvasive procedures.

Medical staff are also more aware that paediatric pro-cedures are often perceived as painful. A survey of doc-tors and nurses working in intermediate or intensive care

J Headache Pain (2006) 7:3–8DOI 10.1007/s10194-006-0267-5

A review of perinatal acute pain: treating perinatal pain to reduce adult chronic pain

R E V I E W

D.A. MarcusDepartment of Anesthesiology,University of Pittsburgh School of Medicine,Pittsburgh, PA, USA

D.A. Marcus (�)Suite 400, Pain Medicine,Centre Commons Building,5750 Centre Avenue,Pittsburgh, PA 15206, USAe-mail: marcusd@anes.upmc.eduTel.: +1-412-953-4797Fax: +1-412-369-8962

Dawn A. Marcus

Abstract Changes in neural con-nections and activity after an acuteinsult are hypothesised to con-tribute to chronic pain syndromesin mature experimental animalsand humans. Over the last decade,studies have suggested that expo-sure to repeated painful proce-dures during the early perinatalperiod results in profound changesin sensitivity of nociceptive path-ways. Both animal and humanstudies show that early pain expe-riences increase pain responsesbeyond the period of infancy.These data suggest a need to

increase implementation of guide-lines for minimising pain expo-sures during infancy. In addition,an experimental perinatal painmodel may provide a uniqueopportunity to study the effects onthe nervous system of both painfulinsults and pre-emptive analgesia.

Keywords Circumcision • Migraine •Neural plasticity • Perinatal pain •

Pre-emptive analgesia

Received: 5 December 2005Accepted in revised form: 16 January 2006Published online: 31 January 2006

4

nurseries showed that 90% believed infants felt the sameor more pain from common procedures than adults [6].Tracheal suctioning, gavage tube insertion and umbilicalcatheter insertion were considered only mildly painful.Moderately painful procedures included intubation, arte-rial or venous cutdown, lumbar puncture, intramuscularinjection, intravenous or intra-arterial line placement,and heel stick. Circumcision and chest tube insertionwere considered to cause severe pain. Despite the aware-ness that these procedures resulted in a pain responsefrom newborns, analgesic measures were not consideredto be routine care. On average, nurses rated all proce-dures as being performed with pharmacologic analgesiconly rarely. Physicians similarly rated all procedures asrarely performed with analgesics, with the exception ofchest tube insertion and cutdowns, which were believedto often be performed with analgesics. Comfort mea-sures were believed to be used substantially more thanpharmacologic treatments, although on average no pro-cedure was rated as usually performed with comfortmeasures. Both nurses and doctors believed that phar-macologic and non-pharmacologic pain-relieving thera-pies should be used in conjunction with painful proce-dures more frequently.

Although the discomfort experienced by earlypainful conditions may be transient, these experiencesmould the nervous system to increase pain sensitivityboth during early development and with maturity. Thisreview will describe both animal and human studies thatdemonstrate the profound impact from early perinatalpainful experiences. These studies support increasedfocus on perinatal pain incidents and the role of neuro-plasticity in chronic pain.

Animal studies

Myelinated A-delta and unmyelinated C fibres transmitnociceptive impulses to the dorsal horn, which acts as animportant sensory processing station and relay centre. Inadult animal studies, peripheral injury results in changesin synaptic transmission within the dorsal horn that mayset the stage for prolonged increased pain sensitivity,resulting in increased risk for chronic pain syndromes.The most widely accepted animal chronic pain model isthe partial sciatic nerve ligation rodent [7, 8]. After expos-ing the sciatic nerve, a temporary ligature is tied aroundthe nerve and later removed. Although the nerve heals andregains neurological function, the rat displays persistentpain behaviours by biting at the previously injured limb.In addition to behavioural changes, autopsy of those ratsdeveloping chronic pain behaviours shows widespread

neurological changes, with re-wiring in the dorsal horn,spine and brain [9, 10]. For example, second-order neu-rons in the dorsal horn show increased sensitivity withincreased action potentials and spontaneous discharges.Central terminals of mechanoreceptors re-distribute with-in the dorsal horn to connect with pain pathway neuronsthat would normally be triggered by pain stimuli.Furthermore, neuronal receptive field size increases.These dorsal horn changes are believed to result in a vari-ety of well reported pain perception changes: increasedsensitivity in the dorsal horn may lead to hyperalgesia;connection of mechanoreceptors to pain pathways mayresult in allodynia; and increased receptive field size mayexplain spread of pain beyond the borders of the initialarea of injury.

A variety of studies have assessed the long-term con-sequences of perinatal pain stimuli in experimental ani-mals. Several studies evaluating adult rats that wereexposed to inflammatory or other noxious stimuli afterbirth are described below. In general, these studies showthat perinatal pain results in increased pain sensitivity inmature rats, which may be at least partially explained byneural plasticity in the dorsal horn. These data suggestthat pain exposure during the critical neonatal periodwhen pain pathways are being developed will lead tolong-standing increased pain sensitivity into adulthood.

Behavioural studies

Anand et al. [11] compared pain responses in adult ratsthat had been exposed from birth to day 7 to stimulation 4times daily with needle sticks or a non-painful tactilestimulus. Tolerance to exposure to cold and hot plate test-ing was performed after the rats matured. Cold testing wassimilar for rats exposed to either needle sticks or tactilestimulus. Hot plate tolerance was significantly lower inrats that had received neonatal needle sticks vs. thosereceiving tactile stimuli when tested at ages 16 days (5.0 svs. 6.2 s; p<0.05) and 22 days (3.9 s vs. 5.5 s; p<0.005). Arecent study failed to confirm these earlier data [12]. Inthis study, newborn rats were similarly exposed to hind-paw needle stick four times daily, however, the controlgroup was also exposed to a similar amount of handlingand maternal separation as the needle stick group. In thisstudy, there was a trend toward the needle stick groupbeing more pain sensitive, with this difference achievingstatistical significance only when differences in maternalgrooming were also considered. These data suggest thatpositive maternal or comfort behaviour could amelioratethe effects of repetitive minor pain exposure in the new-born period.

5

Electrophysiology studies

The impact of neonatal pain on adult neuroanatomy andneurophysiology was tested in an experiment where aninflammatory agent or saline was injected into the hind-paws of newborn rats (days 0–3) [13]. Rats injected withthe inflammation-provoking complete Freund’s adjuvantshowed pain behaviours post-injection and inflammationthat persisted for 5–7 days. Saline-injected rats showedonly immediate withdrawal to needle stick, with no addi-tional pain behaviours or inflammation. Horseradish per-oxidase labelling of dorsal horn primary afferents wasincreased by about 20% in the lower lumbar regions of theinjected side in adult rats who had received neonatalinflammatory injections. Staining for calcitonin gene-related peptide was also increased in rats receivinginflammatory injections as newborn pups. Motoneuronlabelling was similar in both limbs in both groups ofadults. Horseradish peroxidase labelling was also tested inadult rats who had received an inflammatory injection onpostnatal day 14. These rats demonstrated no change inneuronal staining. Possible clinical significance from thisneonatal inflammation was demonstrated with physiolog-ical testing of mature rats, which revealed long-termbehavioural consequences from neonatal pain exposure.Those mature rats that had received neonatal inflammato-ry injections showed an earlier display of the late phase ofpain behaviour after formalin injection. Median time tolate phase was 30.5 min in neonatally treated rats vs. 37.1min in untreated rats (p<0.02), suggesting reduced neuralinhibition or enhanced neural activation. In addition, adultrats exposed to neonatal pain showed increased dorsalhorn neuron firing rates in response to brush and noxiouspinch stimuli compared with untreated rats (p<0.05).

Torsney and Fitzgerald [14] similarly showed persis-tent changes in neurophysiology after neonatal exposureto a full thickness skin flap on the hindpaw. In this study,rat pups were subjected to injection with an inflammatoryagent or skin flap. Extracellular recordings were made inthe L4-5 dorsal horn cells after 3 and 6 weeks. Neithertreatment changed spontaneous activity or mechanicalthresholds. Rats receiving the skin flap injury exhibitedincreased spinal dorsal horn receptive field size by over 2times (p<0.01).

Human studies

In humans, perinatal exposure to severe pain increasespain response in later infancy and adolescence. Taddio etal. [15] compared pain response to routine vaccination inboys at age 4–6 months, based on whether they had been

circumcised or not. In this study, about 70% of all babieshad been circumcised. Demographics were similar for cir-cumcised and uncircumcised babies. After diphtheria-per-tussis-tetanus vaccination, pain ratings were about 50%higher and cry duration was about 3 times longer (22 s vs.7 s) for previously circumcised babies. Similarly, afterHaemophilus influenzae B vaccination, pain scores wereabout 30% higher and cry duration about 3 times longer(53 s vs. 19 s) for previously circumcised babies.

In a later study, this same research group evaluatedpain response to routine injection at 4–6 months in chil-dren who had post-delivery circumcision [16]. All of thebabies had participated in a randomised, placebo-con-trolled clinical trial evaluating the use of Emla anaesthe-sia for circumcision. Pain response 4–6 months later toroutine vaccination was measured by evaluating facialresponses and cry duration, using standardised infant painassessment measures. Demographics and overall tempera-ment were similar in those babies previously circumcisedwith or without anaesthesia. Pain responses, as interpretedby both facial action and cry duration, were about twice ashigh in the infants previously circumcised without anaes-thesia (p<0.05).

Pain sensitivity in adolescents has also been linked toperinatal pain experiences. A case-controlled study com-pared tenderness in 60 adolescents who had been bornprematurely and treated in a neonatal intensive care unit(NICU) with adolescents born full-term [17]. Eighteenareas identified as typically tender in fibromyalgiapatients were used as test locations for tenderness. Inaddition, 9 tender points and 4 control points were alsotested for dolorimetry threshold, using a range of 0–9 kgpressure. Adolescents born prematurely reported signifi-cantly more tender areas (6.0 vs. 3.3; p=0.001) and lowerthreshold for both tender points (4.2 kg vs. 4.8 kg; p=0.04)and control sites (6.3 vs. 7.0; p=0.02). This difference wasmaintained after controlling for gender.

A recent retrospective chart review of paediatricmigraineurs compared childhood migraines in childrentreated as newborns in a NICU or not [18]. As expected,children treated in a NICU experienced more early painprocedures and received more perinatal pain medica-tions. Interestingly, the onset of paediatric migraineoccurred earlier in those previously treated in a NICU(age 7.8 vs. 9.7 years; p<0.01). NICU-treated childrenalso had a greater likelihood of being treated with dailyprophylactic medication for migraine, with preventivetherapies used for about 65% previously treated in aNICU vs. 25% with no NICU exposure. While these datamust be interpreted with caution, they do suggest thatearly painful experiences may change nociceptive path-ways, resulting in earlier onset and more frequent orrecalcitrant migraines.

6

In each of these studies, perinatal pain exposure didnot result in the development of an unprovoked chronicpain syndrome. Early pain, however, does seem toincrease pain sensitivity in those children exposed topainful conditions, including routine vaccinations, tenderpoint assessments or migraine.

Discussion

Even though children and adults do not have consciousmemories of painful events occurring during the perinatalperiod, the nervous system apparently does “remember”those events by altering nociceptive pathways to increasepain sensitivity. Review of animal and human studies onthe effects of perinatal pain supports the need for betterimplementation of analgesic care of newborns, particular-ly those exposed to repeated painful procedures. Thesestudies also suggest that understanding chronic pain syn-dromes in adults may begin by better understanding howearly pain exposures modify nociceptive pathways andpain sensitivity.

The cumulative effect of lifetime pain experiences oncurrent pain perception was evaluated in 49 undergraduatevolunteers [19]. Consistent with gender studies of paintolerance to experimental pain [20], males in this studysimilarly showed a higher pain tolerance, with pain toler-ance time 1.75 times longer in males. In addition, femalesrated their lifetime pain experience level over twice ashigh as the males. A linear relationship was established forboth genders between lifetime pain experience and paintolerance. These data further support that early pain expe-riences may affect later pain perception and tolerance.

Clinical recommendations

Consensus recommendations have been developed toprovide guidelines for analgesic care in newborns [21,22]. First, the staff caring for newborns need to be edu-cated that procedures that would be painful in adults will

also be painful for newborns, including premature babies.Painful procedures should be minimised in newborns. Forexample, venipuncture by trained staff is generally lesspainful than the more commonly performed heel lancing.In addition, line placement may be preferred in newbornsrequiring frequent venous access or injections. Non-inva-sive measures, such as pulse oximetry, should be max-imised to reduce painful monitoring procedures. As inadults, sedation may blunt external pain behaviours innewborns without providing adequate analgesia. Comfortmeasures and analgesia should be considered for neces-sary painful procedures.

A wide variety of analgesic therapies are effective innewborns, including both non-pharmacologic and phar-macologic therapies [23]. Swaddling and sucrose admin-istration via a pacifier are beneficial for most procedures[22, 23]. Both sucking and sucrose administration aloneoffer analgesic benefits, with benefits enhanced by com-bining both together [24]. Other sweet liquids, such asmother’s milk, will also provide analgesic benefit. Topicalanaesthetics should be considered for venipuncture/injec-tions, line placement, lumbar punctures and circumcision[22]. Additional medications, including non-opioid andopioid analgesics may also be considered for other painfulprocedures, such as intubation, suctioning and chest tubeinsertion.

In a recent survey, painful procedures were prospec-tively logged over a 6-month period for all neonates dur-ing their first 14 days of admission in a NICU [25]. Atotal of 151 newborns were followed, with a mean NICUstay of 9 days. Newborns received an average of 14 pro-cedures daily in the NICU. Most procedures were rated asmoderately to severely painful by the treating staff.Despite the frequent exposure to painful procedures, useof analgesia was low. Less than 35% of newbornsreceived analgesics each day. A total of 40% of newbornsreceived no analgesia during their entire NICU stay.Furthermore, failed procedure rates were very high in thissample (Fig. 1), resulting in high exposure to repeatedpainful procedures.

Consensus guidelines recommend implementing edu-cational programmes to increase the skill level of staffperforming procedures on newborns to minimise pain and

Fig. 1 Procedure failure in an NICU(based on Simons et al. [25])

7

the need to repeat failed procedures [21]. The high proce-dure failure rate for a wide range of frequently performedprocedures in the NICU procedure survey [25] supportsthe need to train designated staff to develop an expertisein neonatal procedure administration. Use of comfortmeasures should also be encouraged, particularly in lightof the recent study showing pain sensitivity in mature ratscould be attenuated in newborn pups exposed to repeatedpain with good maternal grooming [12].

The benefits of comfort and analgesia were reported ina brief report evaluating pain response at 8 months of agein low birth-weight babies who had previously been treat-ed in a NICU [26]. Pain response was altered in 8-month-old babies who had been exposed to more invasive proce-dures (p<0.001). In addition, both maternal responsive-ness (p=0.006) and greater use of morphine during theNICU stay (p=0.03) each independently predicted a morenormal pain response at 8 months.

Future research

Better understanding of the neurophysiological conse-quences of pain exposure in the developing nervous sys-tem may broaden our understanding of how the nervoussystem may adapt after trauma or other noxious insult inolder children and adults. In addition, animal perinatalpain models may provide a unique opportunity to evaluatethe effects of pre-emptive analgesia. Pre-emptive analge-sia has been evaluated in several operative models, usingboth opioid and non-opioid analgesia [27–29]. These stud-ies have raised important questions about the best type ofpre-emptive medication (opioid vs. non-opioid), route ofadministration and timing of delivery before painfulinsult. Many of these questions may be carefully studiedusing animal perinatal pain models by evaluating bothanatomical and behavioural consequences of different pre-emptive administrations.

References

1. Rygh LJ, Svendsen F, Fiska A et al(2005) Long-term potentiation inspinal nociceptive systems – how acutepain may become chronic.Psychoneuroendocrinology30:959–964

2. Curatolo M, Arendt-Nielsen L,Petersen-Felix S (2004) Evidence,mechanisms, and clinical implicationsof central hypersensitivity in chronicpain after whiplash injury. Clin J Pain20:469–476

3. Naeini RS, Cahill CM, Ribeiro-da-Silva A, Menard HA, Henry JL (2005)Remodelling of spinal nociceptivemechanisms in an animal model ofmonoarthritis. Eur J Neurosci22:2005–2015

4. Grelik C, Allard S, Ribeiro-da-Silva A(2005) Changes in nociceptive sensoryinnervation in the epidermis of the ratlower lip skin in a model of neuropath-ic pain. Neurosci Lett 389:140–145

5. Porter FL, Wolf CM, Miller JP (1999)Procedural pain in newborn infants:the influence of intensity and develop-ment. Pediatrics 104:e13

6. Porter FL, Wolf CM, Gold J, LotsoffD, Miller JP (1997) Pain and painmanagement in newborn infants: a sur-vey of physicians and nurses.Pediatrics 100:626–632

7. Bennett GJ, Xie YK (1988) A peripher-al mononeuropathy in rat that producesdisorders of pain sensation like thoseseen in man. Pain 33:87–107

8. Seltzer Z, Dubner R, Shir Y (1990) Anovel behavioral model of neuropathicpain disorders produced in rats by par-tial sciatic nerve injury. Pain43:205–218

9. Guilbaud G, Gautron M, Jazat F et al(1993) Time course of degenerationand regeneration of myelinated nervefibers following chronic loose ligaturesof the rat sciatic nerve: can nervelesions be linked to the abnormal pain-related behaviours? Pain 53:147–158

10. Behbehani MM, Dollberg-Stolik O(1994) Partial sciatic nerve ligationresults in an enlargement of the recep-tive field and enhancement of theresponse of dorsal horn neurons tonoxious stimulation by an adenosineagonist. Pain 58:421–428

11. Anand KS, Coskun V, ThrivikramanKV, Nemeroff CB, Plotsky PM (1999)Long-term behavioral effects of repeti-tive pain in neonatal rat pups. PhysiolBehav 66:627–637

12. Johnston CC, Walker C (2003) Theeffects of exposure to repeated minorpain during the neonatal period on for-malin pain behavior and thermal with-drawal latencies. Pain Res Manag8:213–217

13. Ruda MA, Ling QD, Hohmann AG,Pend YB, Tachibana T (2000) Alterednociceptive neuronal circuits afterneonatal peripheral inflammation.Science 289:628–631

14. Torsney C, Fitzgerald M (2003) Spinaldorsal horn cell receptive field size isincreased in adult rats following neona-tal hindpaw skin injury. J Physiol550:255–261

15. Taddio A, Goldbach M, Ipp M, StevensB, Koren G (1995) Effect of neonatalcircumcision on pain responses duringvaccination in boys. Lancet345:291–292

16. Taddio A, Katz J, Ilersich AL, GideonK (1997) Effect of neonatal circumci-sion on pain response during subse-quent routine vaccination. Lancet349:559–603

17. Buskila D, Neumann L, Zmora E,Feldman M, Bolotin A, Press J (2003)Pain sensitivity in prematurely bornadolescents. Arch Pediatr Adolesc Med157:1079–1082

8

18. Maneyapanda SB,Venkatasubramanian A (2005)Relationship between significant peri-natal events and migraine severity.Pediatrics 116:e555–e558

19. Rollman GB, Abdel-Shaheed J,Gillespie JM, Jones KS (2004) Doespast pain influence current pain: bio-logical and psychosocial models of sexdifferences. Eur J Pain 8:427–433

20. Wiesenfeld-Hallin Z (2005) Sex differ-ences in pain perception. Gend Med2:137–145

21. American Academy of Pediatrics.Committee on Fetus and Newborn.Committee on Drugs. Section onAnesthesiology. Section on Surgery.Canadian Paediatric Society. Fetus andNewborn Committee (2000)Prevention and management of painand stress in the neonate. Pediatrics105:454–461

22. Anand KS (2001) Consensus statementfor the prevention and management ofpain in the newborn. Arch PediatrAdolesc Med 155:173–180

23. Anand KS, Johnston CC, OberlanderTF et al (2005) Analgesia and localanesthesia during invasive proceduresin the neonate. Clin Ther 27:844–876

24. Blass EM, Watt LB (1999) Suckling-and sucrose-induced analgesia inhuman newborns. Pain 83:611–623

25. Simons SP, van Dijk M, Anand KS etal (2003) Do we still hurt newbornbabies? A prospective study of proce-dural pain and analgesia in neonates.Arch Pediatr Adolesc Med157:1058–1064

26. Grunau RE (2002) Early pain inpreterm infants. A model of long-termeffects. Clin Perinatol 29:373–394

27. Yukawa Y, Kato F, Ito K, Terashima T,Horie Y (2005) A prospective random-ized study of preemptive analgesia forpostoperative pain in the patientsundergoing posterior lumbar interbodyfusion: continuous subcutaneous mor-phine, continuous epidural morphine,and diclofenac sodium. Spine30:2357–2361

28. Jung YS, Kim MK, Um YJ et al (2005)The effects on postoperative oralsurgery pain by varying NSAID admin-istration times: comparison on effect ofpreemptive analgesia. Oral Surg OralMed Oral Pathol Oral Radiol Endod100:559–563

29. Radhakrishnan M, Bithal PK,Chaturvedi A (2005) Effect of preemp-tive gabapentin on postoperative painrelief and morphine consumption fol-lowing lumbar laminectomy and disce-tomy: a randomized, double-blind,placebo-controlled study. J NeurosurgAnesthesiol 17:125–128