PAIN� 147 (2009) 99–106

w w w . e l se v i e r . c o m / l o c a t e / p a i n

Hindpaw incision in early life increases the hyperalgesic response to repeatsurgical injury: Critical period and dependence on initial afferent activity

Suellen M. Walker a,*, Keri K. Tochiki b, Maria Fitzgerald b

a Portex Unit, Pain Research, UCL Institute of Child Health and Great Ormond St. Hospital, 6th Floor Cardiac Wing, 30 Guilford St., London WC1N 1EH, UKb Neuroscience Physiology and Pharmacology, UCL, Medawar Building, Malet Place, London WC1E 6BT, UK

a r t i c l e i n f o

Article history:Received 23 November 2008Received in revised form 9 July 2009Accepted 18 August 2009

Keywords:Hindpaw incisionPainDevelopmentNeonateHyperalgesiaPrimary afferent

0304-3959/$36.00 � 2009 International Associationdoi:10.1016/j.pain.2009.08.017

* Corresponding author. Tel.: +44(0)20 79052382; fE-mail address: suellen.walker@ich.ucl.ac.uk (S.M.

a b s t r a c t

Pain in early life can enhance the response to subsequent injury, but effects are influenced by both thenature and timing of neonatal injury. Using plantar hindpaw incision, we investigated how postnatalage influences the response to repeat surgical injury two weeks later. The degree and time course ofbehavioural changes in mechanical withdrawal threshold were measured, and injury-related hyperalge-sia was further quantified by flexion reflex electromyographic responses to suprathreshold mechanicalstimuli 24 h following incision. Plantar hindpaw incision produces acute mechanical hyperalgesia in neo-natal and adult rats, but incision in neonatal pups has an additional effect on the response to subsequentinjury. With initial incision at postnatal day (P) 3 or 6, the degree of hyperalgesia following repeat inci-sion 2 weeks later was greater than in animals having a single incision at the same age. At older ages (ini-tial incision at P10, P21 or P40) responses did not differ in repeat and single incision groups. To test therole of primary afferent activity, levobupivacaine sciatic block was performed prior to P6 plantar incision,and controls received saline or subcutaneous levobupivacaine. Repeat peri-operative, but not a singlepre-operative sciatic block, prevented the enhanced response to repeat incision two weeks later. Ourresults show that the first postnatal week represents a critical period when incision increases hyperalge-sia following repeat surgery two weeks later, and effects are initiated by peripheral afferent activity. Thishas potential therapeutic implications for the type and duration of peri-operative analgesia used for neo-natal surgery.

� 2009 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

1. Introduction

Neonates born prematurely and/or with congenital anomaliesoften require surgery in early life, and a proportion will need re-peated procedures or staged repairs throughout childhood. Painand injury related to neonatal surgery occurs at a time when thenormal development of the nervous system is susceptible tochanging levels of neural activity [9]. It is well established that sur-gery can produce persistent pain in adults [24], but the added com-plexity of developmentally-regulated responses to pain and injuryin early life may result in patterns of response and alterations insensory processing that differ from those seen at older ages. Usingquantitative sensory testing, changes in sensory thresholds havebeen demonstrated in children many years following neonatalintensive care [15,17] and neonatal thoracotomy [43], and altera-tions in thermal sensitivity were more marked in those who re-quired surgery in addition to intensive care [51]. Early painexperience has also been shown to increase sensitivity to futureexperimental and clinical stimuli; such as a prolonged thermal

for the Study of Pain. Published by

ax: 44(0)20 78298634.Walker).

stimulus in children who had required neonatal intensive care[15], immunization following neonatal circumcision [47], and re-peat surgery in the same dermatome [32]. Since surgery can be asource of repeated pain experience for patients of all ages, effectsof repeated surgical incision at different stages of neural develop-ment warrant investigation. Furthermore understanding the mech-anisms and developmental periods of susceptibility to injury-induced sensory changes is important for minimising persistent ef-fects of neonatal surgery.

Animal models of neonatal tissue damage indicate that the nat-ure of the injury and the age at which it occurs are both importantdeterminants of subsequent changes in sensory processing. Neona-tal inflammation [8,35], skin wounding [27,36] and nerve injury[18,39] have varying persistent effects with different time win-dows, making it difficult to assess their contribution to the overallimpact of early surgery. Furthermore, previous studies have tendedto use a different ‘test’ pain stimulus from the original ‘priming’one [8,16,35,42,44], which may influence investigation of criticalperiods and underlying mechanisms. In adult rodents, plantarhindpaw incision is an established model of surgical injury [7]and since it also produces acute behavioral hyperalgesia in rat pupsfrom at least one week of age [38,41] we chose it here to further

Elsevier B.V. All rights reserved.

100 S.M. Walker et al. / PAIN� 147 (2009) 99–106

investigate effects of neonatal surgical incision. Hindpaw incisionhas several key advantages. The injury is reproducible and precisespatial coordinates can be identified at all postnatal ages, allowingaccurate comparison of the same stimulus at different ages andtime intervals. Furthermore the restricted anatomical location ofthe injury means that afferent activity at the time of the stimuluscan be effectively manipulated by peripheral nerve blockade.

We have used behavioral and electromyographic (EMG) mea-sures of hindlimb reflex sensitivity to mechanical stimuli followingplantar incision to (i) test the effect of surgical injury at differentpostnatal ages upon the hyperalgesic response to repeat surgerytwo weeks later (ii) identify the developmental time window forthe increased response to repeat injury and (iii) to evaluate the roleof primary afferent activity at the time of initial incision.

2. Methods

2.1. Experimental animals

All experiments were performed under personal and project li-cences in accordance with the United Kingdom Animal (ScientificProcedures) Act 1986. Litters of Sprague–Dawley rat pups were ob-tained from the Biological Services Unit, University College Londonand equal numbers of male and female pups were randomly di-vided into treatment groups. Adult male rats were used for exper-iments commencing after weaning at postnatal day (P) 21. Allanimals were maintained on a 12-h light/dark cycle at constantambient temperature with free access to food and water.

2.2. Plantar hindpaw incision

Animals were anesthetised with halothane (2–4%) in oxygen.Alcoholic chlorhexidine gluconate 0.5% (Vetasept, AnimalcareLtd., York, UK) was applied to the left hindpaw. Following a midlineincision through the skin and fascia of the plantar aspect of thehindpaw, the plantaris muscle was elevated and incised longitudi-nally as previously described [6]. To standardise the relative lengthof the incision at all ages, the incision extended from the midpointof the heel to the proximal border of the first footpad. Skin edgeswere closed with two 5–0 sutures (Ethicon, Edinburgh, UK).

2.3. Behavioral testing

The mechanical threshold for hind-limb withdrawal was deter-mined using calibrated von Frey hairs that deliver a logarithmicallyincreasing force (see Table 1). Pups were habituated to handling,

Table 1Relationship between von Frey hair number and stimulus intensity.

Von Frey hair number Grams

5 0.136 0.227 0.368 0.69 1.0

10 1.611 2.812 4.413 7.814 1315 2216 3717 6018 9019 12020 180

lightly restrained on a flat bench surface, and each von Frey hairwas applied to the dorsal surface of the hindpaw, as this gave sim-ilar but more consistent measures than plantar stimuli in the smallpaws. Adult rats were habituated on an elevated mesh platformand tested on the plantar surface. Von Frey hairs were applied fivetimes at one-second intervals and the number of evoked flexion re-flexes recorded. The maximum force applied was that whichevoked five withdrawal responses [52].

2.4. Electromyography recordings

Electromyography (EMG) recordings were performed as previ-ously described [50,53]. Animals were initially anesthetised withhalothane (2–4%) in oxygen. A tracheostomy was performed andthe animals ventilated with a T-piece system in conjunction witha Harvard small animal ventilator (Harvard Apparatus Ltd.) in atime-cycled, pressure-limited manner. Oxygen was administeredvia bobbin-in-glass rotameters and halothane was delivered froma calibrated vaporiser. Heart rate was continuously monitored(Vetronics ERM-8010 ECG and Respiratory Monitor, Vetronic Ser-vices, Devon, UK) and body temperature was monitored with a rec-tal probe and maintained with a thermostatically controlled heatsource. Animals were placed in a spinal frame with the left hind-paw secured on a fixed platform with the plantar surface exposedfor cutaneous stimulation. Following surgical preparation, the hal-othane concentration was reduced to an age-appropriate concen-tration (from 1.1% at P4 to 0.8% in adults) for 30 min to allowequilibration and establish a stable plane of anaesthesia prior toand during EMG recordings.

Bipolar EMG electrodes (Ainsworks, London) comprising stain-less steel 30G needles with a central copper wire core were placedthrough a small skin incision into the belly of the biceps femorismuscle. EMG responses to mechanical stimuli (von Frey hairs) onthe plantar surface of the hindpaw were processed (Neurolog, Dig-itimer), displayed on a digital storage oscilloscope (Tektronix TDS2012), and recorded in 12-s epochs (PowerLab 4S, AD Instruments,Castle Hill, Australia). As mechanical withdrawal threshold in-creases with age [9,52], more intense mechanical stimuli were re-quired to evoke reflex responses in older animals. Von Frey hairswere sequentially applied to a maximum of hair number 17 (60 gbending force) at P4, hair number 18 (90 g) at P21 and P25, andhair number 20 (180 g) at older ages, resulting in responses beingrecorded to threshold and 2–3 suprathreshold stimuli at each age.As a result, electromyography recordings in anesthetized animalsallow evaluation of changes in threshold as well as in the responseto suprathreshold stimuli, and provide a quantitative measure ofreflex sensitivity and hyperalgesia at a fixed time point followinginjury.

2.5. Experimental plan

Table 2 summarises the experimental groups used for single orrepeat surgeries performed at different ages. Reflex sensitivity wastested using reflex behavioral thresholds and/or EMG recordingspost surgical injury.

Plantar incision was performed in neonatal (postnatal day 3)(n = 4) and adult rats (n = 5) and mechanical withdrawal thresholdswere determined prior to incision and 4, 24, 48 and 72 h later toassess the degree and duration of behavioral mechanical hyperal-gesia. In separate groups of neonatal (P3) and adult (P40) animals,plantar incision was performed, and 24 h later, animals were re-anesthetized for EMG recordings. Quantified reflex responses werecompared with measures from age-matched non-incised controlanimals (n = 4–6 all groups).

To determine if surgery during the neonatal period or adulthoodhad differing effects on the response to repeat incision, rats on

Table 2Experimental outline.

Experiment Age groups Methodology

Behavioural threshold Electromyography recordings

Single incision P3p p

Adult (P40)p p

P3 + P17 versus P17 singlep p

P6 + P20 versus P20 singlep

P10 + P24 versus P24 singlep

P21 + P35 versus P35 singlep

P40 + P54 versus P54 singlep p

Effect of primary afferent block P6 ± sciatic blockp

P6 (± block) + P20 versus P20p

S.M. Walker et al. / PAIN� 147 (2009) 99–106 101

postnatal day (P) 3 or P40 were divided into three treatmentgroups: (i) initial incision and repeat incision 2 weeks later (ii) ini-tial anesthesia only and single incision 2 weeks later and (iii) non-operated age-matched controls (n = 4–5 all groups). Animal groupswere marked and coded to ensure the investigator performingbehavioral testing (KT) was blinded to the single or repeat incisiongrouping. Measurements were performed prior to surgery and atintervals (4, 24, 48 and 72 h, 7 days) post-operatively. In additionalexperiments (as shown in Table 2), reflex sensitivity was quanti-fied with EMG recordings 24 h following single and repeat incisionin neonatal and adult rats (n = 8–10 in all groups). The same inves-tigator (SW) performed all EMG recordings in a standardized man-ner and was blinded to the single or repeat incision grouping.

To determine if there is a critical developmental period duringwhich surgical injury alters the response to repeat injury, initialincision was also performed at P6, P10, or P21 and a second inci-sion was performed 2 weeks later. Twenty-four hours followingincision, EMG recordings quantified the reflex response to mechan-ical stimuli. Values were compared with age-matched single inci-sion and non-incision controls (n = 8–10 in all groups) (see Table2).

2.6. Sciatic nerve block and response to repeat surgery

The effect of primary afferent blockade on acute behavioralmechanical hyperalgesia and on the subsequent response to repeatincision was evaluated by performing incisions in the presence oflocal anesthetic blockade of the sciatic nerve. P6 pups were brieflyanesthetized with halothane and oxygen via a nose cone and per-cutaneous sciatic nerve injections of 50 ll of 0.5% levobupivacaine(Chirocaine, Abbott) were performed using the technique previ-ously described in rat pups [19,25]. Following recovery from anes-thesia, effective sciatic block was confirmed by ipsilateral motorblock and loss of withdrawal reflex response to a suprathresholdmechanical stimulus (vFh 14; 13 g). Pups were then re-anesthe-tized and within 15 min of the block, plantar incision was per-formed lateral to the midline to ensure the wound was withinthe distribution of the sciatic nerve. To compare the effect of dura-

tion of primary afferent blockade, separate groups of pups receivedeither a single pre-incision sciatic block or more prolonged peri-operative blockade. As the size of the pups precludes the use of im-planted catheter techniques, and the duration of sciatic block is rel-atively brief in pups, a pre-operative block plus three percutaneousinjections at 2-h intervals was required to maintain afferent block-ade during the early peri-operative period. Prior to subsequent sci-atic blocks, resolution of motor block and partial recovery fromsensory block was evident. At the same 2-h intervals, littermatecontrol groups received either (i) sciatic injection of 50 ll sterilesaline or (ii) subcutaneous interscapular injection of 50 ll of 0.5%levobupivacaine (n = 6 all groups). Mechanical withdrawal thresh-olds were determined at baseline, post block, and at 4, 24 and 72 hfollowing incision. Two weeks later, a second incision was per-formed and the EMG response quantified 24 h following incision,and compared with P20 pups that had received a single incision(n = 8).

To check that the sciatic nerve block had not affected the nor-mal development of the reflex response, the effect of 4 � 2-h sciaticnerve blocks with levobupivacaine, sciatic injection of saline, orgeneral anesthesia alone were determined at baseline (P6), 24 h(P7) and 2 weeks later (P20). At P21, EMG reflex responses werequantified as previously described (n = 4–6 all groups). To checkthe effects of the nerve block upon the morphological integrity ofthe nerve, repeated P6 injections of saline or levobupivacaine wereperformed as described above (n = 3 both groups) and three dayslater animals were terminally anesthetized with pentobarbitone,transcardially perfused with 4% paraformaldehyde and glutaralde-hyde, and the ipsilateral and contralateral sciatic nerves removed.Ultra-thin sections of the nerves were stained with toluidine blueand examined for structural abnormalities, disruption of the peri-neureum, and infiltration by inflammatory cells.

2.7. Statistical analysis

For each test condition in the behavioral studies, the number ofwithdrawal responses was plotted against the mechanical force ingrams applied by sequential von Frey hairs. As each increase in von

102 S.M. Walker et al. / PAIN� 147 (2009) 99–106

Frey hair number corresponds to a log10 increase in applied force, asigmoidal stimulus response curve with variable slope was con-structed using non-linear regression. The mid point of the curve(50% effective force, EF50) was determined and designated thethreshold [31,52]. Repeated behavioral measures were analysedby two-way ANOVA with time and treatment as variables and Bon-feronni post-tests (GraphPad Prism Version 4).

The duration of the EMG response was outlined from the dis-play of the raw data and the integral of the root mean square(RMS) of the signal was calculated (EMG response) (Chart, Power-lab AD Instruments). The EMG response was plotted against thevon Frey hair number (mechanical stimulus) and the area underthe stimulus–response curve (AUC) calculated to quantify the over-all ‘‘reflex response” (Prism 4, GraphPad, San Diego, USA) [53]. Re-sults were compared with Student’s t-test or one-way ANOVA andpost hoc comparisons depending on the number of experimentalgroups. P < 0.05 was considered statistically significant.

3. Results

3.1. Acute mechanical hyperalgesia following incision in neonatal andadult rats

To confirm that plantar incision in the first postnatal week pro-duces changes in reflex sensitivity, we compared effects in P3 andadult animals. Plantar hindpaw incision produces behavioralmechanical hyperalgesia in both neonatal (Fig. 1A) and adult rats(Fig. 1B) as demonstrated by a leftward shift in the mechanicalstimulus and withdrawal response relationship. In P3 pups, themechanical withdrawal threshold calculated from the midpointof the stimulus–response curve (i.e. EF50) is 1.04 ± 0.05 g at base-line, but is reduced to 0.41 ± 0.03 g at 4 h (P < 0.05; one-way ANO-VA with Dunnett’s post hoc multiple comparison to baseline) and0.80 ± 0.04 g at 24 h following plantar incision. Changes at 24 h

Fig. 1. Acute effects of plantar incision in neonatal and adult animals. The mechanicalresponse (number of withdrawal responses with five applications) at baseline and time p(B; n = 4). The mechanical withdrawal threshold is calculated from the midpoint of thecurve and significant reduction in threshold 4 h following P3 incision, and at all time poimultiple comparison to baseline). Data points = mean ± SEM. (C) Examples of EMG recordcalculating the area under the mechanical stimulus versus EMG response curve (AUC)neonatal (postnatal day 3, P3) or adult (P40) rats, the flexor reflex EMG response (area uincreased when compared to age-matched non-incised controls. Bars = mean ± SEM; n =

are influenced by the normal pattern of increasing mechanicalthreshold with age as demonstrated by the marked right-shift inthe stimulus–response curve 7 days after incision. Withdrawalthresholds at baseline are higher in adult animals (91.1 ± 2.3 g atbaseline) but are significantly reduced to 4.3 ± 1.5 g at 4 h and re-main low for 7 days following incision (P < 0.05 all time points;one-way ANOVA with Dunnett’s multiple comparison to baseline).As previously reported, the duration of mechanical hyperalgesia isshorter in younger animals [41].

To further evaluate reflex sensitivity and hyperalgesia, flexor re-flex electromyography (EMG) recordings were also performed toquantify changes in both threshold and in the response to supra-threshold stimuli (Fig. 1C). Compared to age-matched non-incisedanimals, the EMG reflex response was significantly greater 24 h fol-lowing plantar incision in both neonatal and adult animals(Fig. 1D; P < 0.05 Student’s unpaired two-tailed t-test). Effects didnot differ between male and female pups. Due to developmentaldifferences in reflex responses (increased duration and amplitudeand lower threshold in younger pups) [11,53], the raw value forthe area under the stimulus–response curve is larger in neonatalpups, but incision produces a similar relative increase at both ages(mean 126% increase following P3 incision and 114% increase fol-lowing P40 incision).

3.2. Effect of repeat injury on the degree of mechanical hyperalgesia

To test if the response to repeated surgical injury differs in neo-natal and adult animals, the degree of behavioral mechanicalhyperalgesia following a second incision (prior incision two weeksearlier) was compared with age-matched animals undergoing asingle incision (Fig. 2). The mechanical threshold is reduced inP17 single and repeat incision groups at 4 and 24 h and in adultanimals at all time points to 7 days (P < 0.05 single or repeat inci-sion versus saline, two-way ANOVA with time and treatment as

force applied by sequential von Frey hairs is plotted on a log10 scale against theoints following plantar incision in postnatal day 3 (P3; n = 5) pups (A) and adult ratscurve (i.e. 50% effective force or EF50). There is a left-shift of the stimulus–responsents to 7 days after adult incision (P < 0.05 one-way ANOVA with Dunnett’s post hocings to graded mechanical stimuli (von Frey hairs, vFh 14 to 17) and the method forare shown diagrammatically. (D) Twenty-four hours following plantar incision innder mechanical stimulus versus EMG response curve; AUC EMG) was significantly4–6 all groups. *P < 0.05 Student’s unpaired two-tailed t-test.

Fig. 3. Effect of single and repeat plantar incision at different postnatal ages. Reflexresponse values are derived from the area under the mechanical stimulus versusEMG response curve. At each age, the reflex response value for non-incised controlanimals is standardized to 100%. The change in reflex response following single orrepeat incision is expressed as a percentage change from non-incised controls. Theage at time of incision is shown for animals having a repeat incision with priorincision two weeks earlier, and age-matched animals having a single incision. Datapoints = mean ± SEM; n = 8–10 all groups; **P < 0.01 single or repeat incision versuscontrol; §§P < 0.01 single versus repeat, two-way ANOVA with Bonferroni post hoccomparison.

S.M. Walker et al. / PAIN� 147 (2009) 99–106 103

variables and Bonferroni post hoc comparison). There is a main ef-fect of treatment (neonatal group F(2,66) = 33.5, P < 0.001; adultgroup F(2,54) = 1029, P < 0.0001), and a significant interaction withtime (neonatal group F(10,66) = 31.4, P < 0.001; adult groupF(10,54) = 49.8, P < 0.001). The increase in reflex thresholds in neona-tal animals at 72 h (now aged P6) and 7 days (now aged P10) isconsistent with normal developmental changes in the reflex re-sponse. Due to these changes in baseline with age, time accountsfor more variance in the neonatal group (82% versus 25%) and fur-ther analysis compared the response to single versus repeat inci-sion within the separate age groups. When the first incision wasperformed as a neonate (P3), the decrease in mechanical threshold24 and 48 h following a second incision two weeks later was signif-icantly greater than in animals having a single incision at this age(Fig. 2A; P < 0.05 single versus repeat incision, two-way ANOVAwith time and treatment as variables and Bonferroni post hoc com-parison). When the first incision was performed as an adult (P40),the degree of behavioral mechanical hyperalgesia following thesecond injury two weeks later was the same as that following asingle injury at this age (Fig. 2B).

In neonatal and adult animals, the hyperalgesic response to re-peat injury was also measured by flexion reflex EMG responses tograded von Frey hair mechanical stimuli (Fig. 3). As the baseline re-flex response changes with age (as seen in Fig. 1D), the value fornon-incised controls was normalized to 100% at all ages, and theeffect of single or repeat incision expressed as a percentage changefrom control (% change reflex response = [(incision AUC � controlAUC)/control AUC] � 100). Consistent with the behavioral mea-sures, age-related differences in the response to single and repeat

Fig. 2. Mechanical withdrawal thresholds following single or repeat hindpawincision in neonatal and adult rats. Mechanical withdrawal thresholds (EF50 ingrams) at baseline and subsequent time points are shown following repeat incision(initial P3 and second incision 2 weeks later; P3 + P17 repeat) and compared withsingle incision at P17 (P17 single) and age-matched non-incised control animals(Fig. 2A). Fig. 2B shows mechanical thresholds in adult animals following repeatincision (P40 and P54 repeat) or single incision (P54 single). Controls measures arefrom the left hindpaw of age-matched, non-operated animals. Data points = -mean ± SEM; n = 4–5 all groups; §P < 0.05 single versus repeat; +P < 0.05 repeatversus control; *P < 0.05 single versus control; two-way ANOVA with time andtreatment as variables and Bonferroni post hoc comparison.

incision were noted. Prior incision in neonatal but not in adult ani-mals was associated with increased hyperalgesia following a sec-ond incision two weeks later, as the reflex response in the repeatincision group was significantly greater than the single incisiongroup (P < 0.01 single versus repeat, two-way ANOVA with Bonfer-roni post hoc comparison) (Fig. 3).

To test the critical age period for producing this enhanced re-sponse to repeat injury, initial plantar hindpaw incision was alsoperformed at P6, P10, or P21 and the reflex hyperalgesia followinga second incision 2 weeks later was tested. At all ages, plantar inci-sion produced hyperalgesia that was quantifiable with EMGrecordings. There was a main effect of treatment (F(2,121) = 217.6,P < 0.0001) which accounted for 66% of the total variance, and aninteraction with age (F(8,121) = 10.3, P < 0.01) which accounted for11% of the total variance. Twenty-four hours following incision,the reflex response was significantly greater in both the singleand repeat incision groups at all ages (P < 0.01 single or repeat inci-sion versus non-incised control, two-way ANOVA with Bonferronipost hoc comparison). As seen at P3, prior incision at P6 was asso-ciated with an enhanced hyperalgesic response to a second incisiontwo weeks later (Fig. 3; P < 0.01 single versus repeat incision, two-way ANOVA with Bonferroni post hoc comparison), but when thefirst incision was performed at older ages (i.e. P10, P21 or P40)the response to repeat incision did not differ from age-matchedsingle incision groups.

3.3. Effect of primary afferent blockade at the time of initial surgery

To determine if blocking primary afferent activity modifiesacute mechanical hyperalgesia following neonatal incision, the ef-fects of pre-operative or more prolonged peri-operative sciaticblockade on mechanical hindpaw withdrawal thresholds beforeand after P6 incision were measured (Fig. 4A). As expected,mechanical hyperalgesia was apparent in the saline group as themechanical withdrawal threshold of the incised paw was signifi-cantly decreased from baseline 4 and 24 h following incision(P < 0.01 one way repeated measures ANOVA with Bonferroni posthoc comparison). In addition, the threshold in the saline group waslower when compared to both sciatic block groups at 4 and 24 h(P < 0.05 saline versus pre- or peri-operative sciatic block, twoway repeated measures ANOVA and Bonferroni post hoc compari-

Fig. 4. Effect of sciatic nerve block on the acute behavioral response to plantarincision at P6 and the response to a second incision. (A) Behavioral mechanicalwithdrawal thresholds of the ipsilateral hindpaw in P6 pups at baseline and 4, 24and 72 h following plantar incision are plotted. Treatment groups include: a singlepre-operative injection of levobupivacaine (pre-op sciatic block 1�) or a pre-operative block plus 3 � 2-h post-operative blocks (peri-operative sciatic blocks4�). Control animals received 4 � 2 h injections of saline (saline). Bars = mean ± -SEM; n = 7–8. *P < 0.05 saline versus pre-operative sciatic (1�) and versus peri-operative sciatic (4�); two-way ANOVA with time and treatment as variables andBonferroni post hoc comparison. (B) The reflex response following single incision atP20 (single incision) is compared with P6 + P20 incision (repeat incision) in animalstreated at the time of P6 incision with: 4 � 2-h injections of saline (saline); 4 � 2-hsubcutaneous injections of the local anesthetic levobupivacaine (s.c. LA); a singlepre-operative levobupivacaine sciatic block (pre-operative sciatic 1�); or 4x2-hlevobupivacaine sciatic blocks (peri-operative sciatic 4�). The reflex response torepeat incision was significantly greater compared with the single incision group inanimals treated with saline, subcutaneous local anesthetic or a single pre-operativeblock, but the repeated peri-operative sciatic blocks did not differ from the singleincision group. AUC EMG = area under mechanical stimulus versus EMG–responsecurve; Bars = mean ± SEM; n = 6–8 all groups; *P < 0.05 one-way ANOVA withDunnett’s multiple comparison to single incision group.

104 S.M. Walker et al. / PAIN� 147 (2009) 99–106

son). Hyperalgesia was not apparent in either the single pre-oper-ative or prolonged peri-operative sciatic block groups, as themechanical threshold did not differ significantly from baseline ateither 4 or 24 h post incision (P > 0.05 two way repeated measuresANOVA and Bonferroni post hoc comparison). By 72 h, there is anage-related increase in mechanical threshold but incision-relatedmechanical hyperalgesia has resolved, as there was no significantdifference in threshold across treatment groups (Fig. 4A, P > 0.05one-way ANOVA and Bonferroni post hoc comparison), or betweenthe ipsilateral and contralateral paw in any group at this time (datanot shown).

To determine if primary afferent activity at the time of initialincision contributes to subsequent sensitivity, we also tested theeffect of sciatic block at the time of initial P6 incision upon thehyperalgesic response to repeat incision 2 weeks later. Fig. 4Bshows that peri-operative sciatic nerve blockade prevented the en-hanced response to subsequent injury as the quantified reflex re-sponse following repeat incision did not differ from animalshaving a single incision at this age. Animals treated with only a sin-gle pre-operative sciatic nerve block at the time of initial P6 inci-sion, and controls receiving repeated subcutaneous injections ofthe same dose of levobupivacaine or sciatic injections of saline,

all demonstrated an enhanced response to repeat incision, as thereflex response was significantly greater than the single incisiongroup (Fig. 4B, P < 0.05 one-way ANOVA with Dunnett’s post hoccomparison to single incision group).

The effect of repeated sciatic block was restricted to the re-sponse to repeat injury; there was no effect upon on the normaldevelopment of the flexion reflex response. When comparing ani-mals receiving anesthesia plus repeated (4 � 2-h) sciatic injectionsof levobupivacaine, or anesthesia plus saline injection, or repeatedanesthesia alone, there were no differences in weight gain at 24 hor 2 weeks (mean ± SEM: 41 ± 0.9 versus 41 ± 1.2 g versus43 ± 0.9 g, respectively at P20, P > 0.05 one-way ANOVA with Bon-ferroni post hoc comparison) or behavioral mechanical withdrawalthresholds at 24 h or 2 weeks (mean 6–7 g all groups at P20), or inthe quantified EMG response at P21 (mean ± SEM: 012 ± 0.03 ver-sus 0.10 ± 0.02 versus 0.11 ± 0.05, respectively, P > 0.05 one-wayANOVA with Bonferroni post hoc comparison). Repeat injectionof levobupivacaine or saline produced no light microscopic struc-tural differences visible in ultra-thin sections. There was no disrup-tion of the perineurium or inflammatory cell infiltrate in the sciaticnerve following repeated injection of levobupivacaine (Supple-mentary Fig. 1) or saline (Supplementary Fig. 2), and there wereno differences from contralateral non-injected nerves.

4. Discussion

Our results show that the hyperalgesic response to repeat sur-gery after a two-week interval is enhanced by prior neonatal expe-rience, and this effect is dependent on the age at which the initialinjury occurs. Plantar hindpaw incision produces acute hyperalge-sia in neonatal and adult rats, but incision in neonatal pups has anadditional effect on the response to subsequent injury, that is notseen following initial incision at older ages. The first postnatalweek represents a critical period during which incision producesa segmental hyperalgesia that is unmasked by repeat injury twoweeks later. This effect is prevented by peri-operative sciatic nerveblockade suggesting a mechanistic role for primary afferent activ-ity at the time of initial incision.

Plantar hindpaw incision is an established post-operative painmodel in adult rats [4,6], and the mechanisms underlying hyperal-gesia and the efficacy of analgesics differ from inflammation andnerve injury [7,28]. Age-related effects of surgical injury warrantspecific evaluation, and using anatomical hindpaw landmarks toinjure the same relative area ensures a consistent and reproduciblestimulus despite changes in size with age. Consistent with datafrom P14 pups [41], a single plantar incision at P3 or P17 producedbehavioral mechanical hyperalgesia, but the duration was shorterthan in adults. We also quantified sensitivity by flexion reflexEMG recordings to assess changes in threshold and in the responseto suprathreshold stimuli, both of which contribute to injury-re-lated hyperalgesia. We have previously used this methodology toevaluate C-fiber induced mechanical hyperalgesia in rat pups[53] and have now confirmed that plantar incision produces in-jury-related hyperalgesia at all postnatal ages.

Different forms of neonatal injury have been associated withalterations in the response to future noxious stimuli. Severe CFA-induced neonatal inflammation increases the response to inflam-mation, formalin, and capsaicin injection of the previously injuredpaw in adulthood [16,42,46]. Mild neonatal inflammation fromhindpaw carrageenan is associated with increased hyperalgesiawhen the ipsilateral, but not contralateral paw is re-inflamed[35] or surgically incised [8] in adulthood. As surgery produces skinwounding and nerve injury in addition to inflammation [24,55],but the developmental profile of these different forms of injuryvaries [18,27,36,39], we specifically evaluated the impact of hind-

S.M. Walker et al. / PAIN� 147 (2009) 99–106 105

paw incision in early life upon the pain evoked by repeat surgicalinjury. Initial incision at P3 was associated with increased mechan-ical hyperalgesia (as assessed by behavioral withdrawal thresholdsand EMG measures of reflex sensitivity) following a repeat incisiontwo weeks later, whereas prior experience had no impact when theinitial incision occurred in adulthood.

A developmental window or critical period during which injurytriggers persistent sensory changes in rodents has been reportedfollowing inflammation. Severe inflammation at P0 or P3 but notat P14 produces long-term expansion of sciatic terminal fields inthe spinal cord [42]; mild inflammation at P3 but not P14 is asso-ciated with alterations in visceral sensitivity in the adult [54]; andan enhanced response to repeat inflammation is seen if mildinflammation is induced between P0 and P5, but not at P8 or older[35]. Surgical laparotomy in P0 mice produces long-term sensorychanges with reduced sensitivity to visceral stimuli in adulthood[44], but segmental effects of repeat incision have not been evalu-ated in this model. By comparing effects of surgical injury across arange of ages, we have now shown that the first postnatal week is acritical period for altered responses to surgery, as hindpaw incisionat P3 or P6, but not at older ages, was associated with increasedhyperalgesia following repeat incision two weeks later. An en-hanced response to re-inflammation of a neonatally-inflamedpaw is apparent at time points from 4 days until several weeksafter the initial injury [35]. Repeat incision was performed after2 weeks to allow complete wound healing and resolution of pri-mary hyperalgesia at all ages, so the exact time of onset ofenhancement is not clear. We are currently evaluating effects ofneonatal incision on sensory processing in adulthood [48].

As the enhanced response to repeat surgery occurs only afterneonatal incision, the injury must trigger plastic changes in painpathways through developmentally regulated mechanisms thatdiffer from those following adult plantar incision. The efficacy ofperi-operative sciatic block indicates a crucial initiating role forprimary afferent activity arising from the skin incision. Such activ-ity may cause local changes in sensory terminals, such as hyperin-nervation [27] or alterations in receptor transport and distribution[21], or central effects such as sensitization of nociceptive circuitsin the spinal cord. Plantar incision in P7 and P28 pups increasesexcitability in large DRG neurons, but ongoing activity persists be-yond the period of behavioral hyperalgesia in the younger pupsand may contribute to longer-term changes [40]. In the adult, plan-tar incision produces spontaneous activity in A-beta as well as A-delta and C primary afferent fibers [13,28,33], in a pattern that dif-fers from inflammation and nerve injury [26,29]. As the central dis-tribution and functional response to C- and A-fiber afferent inputschange during the postnatal period [2,20,30,53], and A-beta fiberactivation produces sensitization in the superficial dorsal horn inyoung but not adult animals [20,30], activation of A-beta fibresby incision may contribute to differing effects in earlydevelopment.

Plantar incision produces alterations in the stimulus–responsecharacteristics of dorsal horn neurons and central sensitization inadult rodents [23,34,49]. In P7 pups, greater evoked spike activityand a rapid expansion of receptive field size are seen in the dorsalhorn [38]. As there are significant postnatal changes in excitatoryand inhibitory circuits in the spinal cord and in their modulationby descending pathways [1,9,14], activity-dependent alterationsin nociceptive circuitry during this period may underlie persistentfunctional changes. c-Fos expression in the dorsal horn is increasedipsilateral to repeat inflammation [46], capsaicin injection [16] orplantar incision [8] when a neonatally-inflamed paw is re-injuredin adulthood. However, differences in intracellular signalling [8]suggest that it is not just the degree, but also the mechanisms ofcentral sensitivity that are altered by early injury.

The duration of primary afferent blockade was critical formodulating the response to neonatal plantar incision. Two differ-ent patterns of response were identified: (i) a reduction inbehavioral mechanical hyperalgesia that persisted beyond theduration of sciatic blockade (i.e. preventive analgesic effect[22]), and (ii) modulation of the enhanced response to repeatincision two weeks later. In adult rats, single pre-incision admin-istration of local anesthetic (either peripheral infiltration [34],ankle blocks plus plantar infiltration [45], or intrathecal injection[5]) reduced behavioral mechanical hyperalgesia at 4 h, but not24 h. This lack of preventive effects in adults may be influencedby the longer duration of hyperalgesia relative to the local anes-thetic block. By contrast, in younger pups a single pre-incisionsciatic block (seen here in P6 pups and previously reported atP14 [37]) does reduce mechanical hyperalgesia 24 h followinghindpaw incision. More prolonged peri-operative sciatic blockadeat P6 had an additional impact on the response to a second inci-sion two weeks later that was not seen following a single pre-operative block, suggesting that the duration of an analgesicintervention may be critical for modulating more persistent ef-fects following neonatal surgery.

As systemic administration of local anesthetic can produceanalgesic effects following hindpaw incision [23], effects ofperipheral nerve blockade need to be differentiated from effectsof systemic absorption [3]. In the current experiments, sciaticbut not subcutaneous administration of levobupivacaine pre-vented the enhanced response to repeat incision, suggesting aspecific role for primary afferent blockade. The reduction in re-flex response following repeat incision was not due to impaireddevelopment of the reflex, as repeated sciatic blocks (in the ab-sence of surgical injury) had no effect on behavioral thresholdsor EMG responses, and produced no histological signs of sciaticnerve injury.

Our findings are clinically relevant from two different perspec-tives. Firstly, increased peri-operative analgesic requirements atthe time of repeat surgery has been shown in infants, but similarstudies have not been performed following initial surgery at olderages [32]. The current data suggests that the enhanced hyperalge-sic response to repeat incision after two weeks specifically relatesto an initial injury in the neonatal period. While direct parallels be-tween rat and human postnatal development are difficult, spinalcord development during the first postnatal week in rodents corre-lates with the human preterm and neonatal period [10,12], sug-gesting that initial surgery during this time may be associatedwith age-related changes in sensory processing. Secondly, our find-ings have implications for the ability of analgesia at the time of theinitial injury to modulate longer-term alterations in pain. Topicallocal anesthesia at the time of neonatal circumcision reduced butdid not completely prevent the enhanced behavioral response toimmunization several months later [47]. Here we find that the abil-ity to prevent subsequent enhanced hyperalgesia was dependenton the duration of the initial peripheral nerve blockade. This dataprovides important insights into the plastic changes in pain pro-cessing triggered by early pain experience and has potential ther-apeutic implications for the type and duration of peri-operativeanalgesia used for neonatal surgery.

Acknowledgments

This research was funded by the International Association forthe Study of Pain (IASP Early Career Research Grant, SW) and theMedical Research Council of Great Britain (MRC, MF).

The authors declare that they have no conflicts of interest,including specific financial interests, relationships or affiliationsrelevant to the manuscript.

106 S.M. Walker et al. / PAIN� 147 (2009) 99–106

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.pain.2009.08.017.

References

[1] Baccei ML. Development of pain: maturation of spinal inhibitory networks. IntAnesthesiol Clin 2007;45:1–11.

[2] Baccei ML, Bardoni R, Fitzgerald M. Development of nociceptive synapticinputs to the neonatal rat dorsal horn: glutamate release by capsaicin andmenthol. J Physiol 2003;549:231–42.

[3] Beloeil H, Ababneh Z, Chung R, Zurakowski D, Mulkern RV, Berde CB. Effects ofbupivacaine and tetrodotoxin on carrageenan-induced hind paw inflammationin rats (part 1): hyperalgesia, edema, and systemic cytokines. Anesthesiology2006;105:128–38.

[4] Brennan TJ. Incisional sensitivity and pain measurements: dissectingmechanisms for postoperative pain. Anesthesiology 2005;103:3–4.

[5] Brennan TJ, Umali EF, Zahn PK. Comparison of pre- versus post-incisionadministration of intrathecal bupivacaine and intrathecal morphine in a ratmodel of postoperative pain. Anesthesiology 1997;87:1517–28.

[6] Brennan TJ, Vandermeulen EP, Gebhart GF. Characterization of a rat model ofincisional pain. Pain 1996;64:493–501.

[7] Brennan TJ, Zahn PK, Pogatzki-Zahn EM. Mechanisms of incisional pain.Anesthesiol Clin North America 2005;23:1–20.

[8] Chu YC, Chan KH, Tsou MY, Lin SM, Hsieh YC, Tao YX. Mechanical painhypersensitivity after incisional surgery is enhanced in rats subjected toneonatal peripheral inflammation: effects of N-methyl-D-aspartate receptorantagonists. Anesthesiology 2007;106:1204–12.

[9] Fitzgerald M. The development of nociceptive circuits. Nat Rev Neurosci2005;6:507–20.

[10] Fitzgerald M. The developmental neurobiology of pain. In: Bond M, Charlton JE,Woolf CJ, editors. Proceedings of the VIth World congress on pain research andclinical management, Vol. 4. Amsterdam: Elsevier; 1991. p. 253–61.

[11] Fitzgerald M, Gibson S. The postnatal physiological and neurochemicaldevelopment of peripheral sensory C fibres. Neuroscience 1984;13:933–44.

[12] Fitzgerald M, Shaw A, MacIntosh N. Postnatal development of the cutaneousflexor reflex: comparative study of preterm infants and newborn rat pups. DevMed Child Neurol 1988;30:520–6.

[13] Hamalainen MM, Gebhart GF, Brennan TJ. Acute effect of an incision onmechanosensitive afferents in the plantar rat hindpaw. J Neurophysiol2002;87:712–20.

[14] Hathway G, Harrop E, Baccei M, Walker S, Moss A, Fitzgerald M. A postnatalswitch in GABAergic control of spinal cutaneous reflexes. Eur J Neurosci2006;23:112–8.

[15] Hermann C, Hohmeister J, Demirakca S, Zohsel K, Flor H. Long-term alterationof pain sensitivity in school-aged children with early pain experiences. Pain2006;125:278–85.

[16] Hohmann AG, Neely MH, Pina J, Nackley AG. Neonatal chronic hind pawinflammation alters sensitization to intradermal capsaicin in adult rats: abehavioral and immunocytochemical study. J Pain 2005;6:798–808.

[17] Hohmeister J, Demirakca S, Zohsel K, Flor H, Hermann C. Responses to pain inschool-aged children with experience in a neonatal intensive care unit:cognitive aspects and maternal influences. Eur J Pain 2009;13:94–101.

[18] Howard RF, Walker SM, Mota PM, Fitzgerald M. The ontogeny of neuropathicpain: postnatal onset of mechanical allodynia in rat spared nerve injury (SNI)and chronic constriction injury (CCI) models. Pain 2005;115:382–9.

[19] Hu D, Hu R, Berde CB. Neurologic evaluation of infant and adult rats before andafter sciatic nerve blockade. Anesthesiology 1997;86:957–65.

[20] Jennings E, Fitzgerald M. Postnatal changes in responses of rat dorsal horn cellsto afferent stimulation: a fibre-induced sensitization. J Physiol1998;509:859–68.

[21] Ji RR, Samad TA, Jin SX, Schmoll R, Woolf CJ. P38 MAPK activation by NGF inprimary sensory neurons after inflammation increases TRPV1 levels andmaintains heat hyperalgesia. Neuron 2002;36:57–68.

[22] Katz J, McCartney CJ. Current status of preemptive analgesia. Curr OpinAnaesthesiol 2002;15:435–41.

[23] Kawamata M, Furue H, Kozuka Y, Narimatsu E, Yoshimura M, Namiki A.Changes in properties of substantia gelatinosa neurons after surgical incisionin the rat: in vivo patch-clamp analysis. Anesthesiology 2006;104:432–40.

[24] Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors andprevention. Lancet 2006;367:1618–25.

[25] Kohane DS, Sankar WN, Shubina M, Hu D, Rifai N, Berde CB. Sciatic nerveblockade in infant, adolescent, and adult rats: a comparison of ropivacainewith bupivacaine. Anesthesiology 1998;89:1199–208.

[26] Matsumoto M, Inoue M, Hald A, Yamaguchi A, Ueda H. Characterization ofthree different sensory fibers by use of neonatal capsaicin treatment, spinalantagonism and a novel electrical stimulation-induced paw flexion test. MolPain 2006;2:16.

[27] Moss A, Alvares D, Meredith-Middleton J, Robinson M, Slater R, Hunt SP,Fitzgerald M. Ephrin-A4 inhibits sensory neurite outgrowth and is regulatedby neonatal skin wounding. Eur J Neurosci 2005;22:2413–21.

[28] Nagakura Y, Jones TL, Malkmus SA, Sorkin L, Yaksh TL. The sensitization of abroad spectrum of sensory nerve fibers in a rat model of acute postoperativepain and its response to intrathecal pharmacotherapy. Pain 2008;139:569–77.

[29] Nagakura Y, Malkmus S, Yaksh TL. Determination of current threshold for pawwithdrawal with sine-wave electrical stimulation in rats: effect of drugs andalteration in acute inflammation. Pain 2008;134:293–301.

[30] Nakatsuka T, Ataka T, Kumamoto E, Tamaki T, Yoshimura M. Alteration insynaptic inputs through C-afferent fibers to substantia gelatinosa neurons ofthe rat spinal dorsal horn during postnatal development. Neuroscience2000;99:549–56.

[31] Nandi R, Beacham D, Middleton J, Koltzenburg M, Howard RF, Fitzgerald M.The functional expression of mu opioid receptors on sensory neurons isdevelopmentally regulated: morphine analgesia is less selective in theneonate. Pain 2004;111:38–50.

[32] Peters JW, Schouw R, Anand KJ, van Dijk M, Duivenvoorden HJ, Tibboel D. Doesneonatal surgery lead to increased pain sensitivity in later childhood? Pain2005;114:444–54.

[33] Pogatzki EM, Gebhart GF, Brennan TJ. Characterization of Adelta- and C-fibersinnervating the plantar rat hindpaw one day after an incision. J Neurophysiol2002;87:721–31.

[34] Pogatzki EM, Vandermeulen EP, Brennan TJ. Effect of plantar local anestheticinjection on dorsal horn neuron activity and pain behaviors caused by incision.Pain 2002;97:151–61.

[35] Ren K, Anseloni V, Zou SP, Wade EB, Novikova SI, Ennis M, Traub RJ, Gold MS,Dubner R, Lidow MS. Characterization of basal and re-inflammation-associatedlong-term alteration in pain responsivity following short-lasting neonatal localinflammatory insult. Pain 2004;110:588–96.

[36] Reynolds ML, Fitzgerald M. Long-term sensory hyperinnervation followingneonatal skin wounds. J Comp Neurol 1995;358:487–98.

[37] Ririe DG, Barclay D, Prout H, Tong C, Tobin JR, Eisenach JC. Preoperative sciaticnerve block decreases mechanical allodynia more in young rats: is preemptiveanalgesia developmentally modulated? Anesth Analg 2004;99:140–5.

[38] Ririe DG, Bremner LR, Fitzgerald M. Comparison of the immediate effects ofsurgical incision on dorsal horn neuronal receptive field size and responsesduring postnatal development. Anesthesiology 2008;109:698–706.

[39] Ririe DG, Eisenach JC. Age-dependent responses to nerve injury-inducedmechanical allodynia. Anesthesiology 2006;104:344–50.

[40] Ririe DG, Liu B, Clayton B, Tong C, Eisenach JC. Electrophysiologiccharacteristics of large neurons in dorsal root ganglia during developmentand after hind paw incision in the rat. Anesthesiology 2008;109:111–7.

[41] Ririe DG, Vernon TL, Tobin JR, Eisenach JC. Age-dependent responses tothermal hyperalgesia and mechanical allodynia in a rat model of acutepostoperative pain. Anesthesiology 2003;99:443–8.

[42] Ruda MA, Ling QD, Hohmann AG, Peng YB, Tachibana T. Altered nociceptiveneuronal circuits after neonatal peripheral inflammation. Science2000;289:628–31.

[43] Schmelzle-Lubiecki BM, Campbell KA, Howard RH, Franck L, Fitzgerald M.Long-term consequences of early infant injury and trauma uponsomatosensory processing. Eur J Pain 2007;11:799–809.

[44] Sternberg WF, Scorr L, Smith LD, Ridgway CG, Stout M. Long-term effects ofneonatal surgery on adulthood pain behavior. Pain 2005;113:347–53.

[45] Sun X, Yokoyama M, Mizobuchi S, Kaku R, Nakatsuka H, Takahashi T, Morita K.The effects of pretreatment with lidocaine or bupivacaine on the spatial andtemporal expression of c-Fos protein in the spinal cord caused by plantarincision in the rat. Anesth Analg 2004;98:1093–8.

[46] Tachibana T, Ling QD, Ruda MA. Increased Fos induction in adult rats thatexperienced neonatal peripheral inflammation. Neuroreport 2001;12:925–7.

[47] Taddio A, Katz J, Ilersich AL, Koren G. Effect of neonatal circumcision on painresponse during subsequent routine vaccination. Lancet 1997;349:599–603.

[48] Tochiki K, Beggs S, Fitzgerald M, Walker SM. Neonatal rat hindpaw incisionand enhanced pain sensitivity in later life: role of afferent activity and spinalmicroglia. Abstracts of IASP 12th World Congress on Pain 2008:PW283.

[49] Vandermeulen EP, Brennan TJ. Alterations in ascending dorsal horn neurons bya surgical incision in the rat foot. Anesthesiology 2000;93:1294–302.

[50] Walker SM, Fitzgerald M. Characterization of spinal alpha-adrenergicmodulation of nociceptive transmission and hyperalgesia throughoutpostnatal development in rats. Br J Pharmacol 2007;151:1334–42.

[51] Walker SM, Franck LS, Fitzgerald M, Myles J, Stocks J, Marlow N. Long-termimpact of neonatal intensive care and surgery on somatosensory perception inchildren born extremely preterm. Pain 2009;141:79–87.

[52] Walker SM, Howard RF, Keay KA, Fitzgerald M. Developmental age influencesthe effect of epidural dexmedetomidine on inflammatory hyperalgesia in ratpups. Anesthesiology 2005;102:1226–34.

[53] Walker SM, Meredith-Middleton J, Lickiss T, Moss A, Fitzgerald M. Primary andsecondary hyperalgesia can be differentiated by postnatal age and ERKactivation in the spinal dorsal horn of the rat pup. Pain 2007;128:157–68.

[54] Wang G, Ji Y, Lidow MS, Traub RJ. Neonatal hind paw injury alters processingof visceral and somatic nociceptive stimuli in the adult rat. J Pain2004;5:440–9.

[55] Whiteside GT, Harrison J, Boulet J, Mark L, Pearson M, Gottshall S, Walker K.Pharmacological characterisation of a rat model of incisional pain. Br JPharmacol 2004;141:85–91.