Il RTrA

y,

d O

Funinjurycisely.nervecoverition inand mand bmotorWe deing mmentsrecovetranse

Keytesting

Periteredfunctidevelolowingregeneand min thelogic,regeneration have all been established (2, 57, 11, 12).Of these, functional recovery is the most difficult tomeasuwith h

Thecoveryfunctinaceli

(1)thin

ianla ivesirrnalghanallectreatry

pleve

wnewsthingen

mse

Inbred adult male Fisher rats (n 5 8), 150200 g,were used. All animals were housed three per cage

1 To wOtolaryInfirma7846. E

Experimental Neurology 168, 192195 (2001)doi:10.1006/exnr.2000.7600, available online at http://www.idealibrary.com on

0014-488CopyrighAll rightre precisely and does not always correlate wellistologic and electrophysiologic results (13).current standard for measuring functional re-following rat sciatic nerve injury is the sciatic

on index (SFI), established in 1982 by DeMedi-et al. (6) and subsequently modified by Bain et

with a 12-h light/dark cycle, and were fed rat chow andwater ad libitum, following Animal Care and Use Com-mittee guidelines. The animals were handled on a dailybasis for a 2-week period prior to the study, to accli-mate them with the testing area and procedure, and tominimize anxiety-related testing inaccuracies (14).

Following anesthetic induction using inhalationalmethoxyfluorane, the left hindlimb was shaved andsterilely prepped. The sciatic nerve was exposed at thehom correspondence should be addressed at Department ofngology/Head and Neck Surgery, Massachusetts Eye and EarBRIEF COMMUN

Quantification of FunctionaRat Sciatic Nerve

Rahul Koka and TessaDepartment of Otolaryngology, Massachusetts Eye and Ear Infirmar

Received May 5, 2000; accepte

ctional recovery following experimental nervehas been notoriously difficult to quantify pre-The current gold standard in the rat sciatic

model involves analysis of footprints of the re-ng animal, and computation of the sciatic func-dex (SFI). We performed transection injuries

easured recovery both by walking track analysisy a newer, simpler, more quantitative test ofrecovery, the extensor postural thrust (EPT).

monstrate a high correlation between both test-odalities and suggest a role for EPT measure-as an easier, more consistent measure of motorry following experimental rat sciatic nervection. 2001 Academic PressWords: rat sciatic nerve; transection; behavioral.

pheral nerve injuries are a commonly encoun-clinical problem and often result in long-termonal deficits. Extensive investigation toward thepment of methods to improve regeneration fol-nerve injury is ongoing. Attempts to study the

rative effect of various neurotrophic substancesanipulations have been complicated by difficulty

precise quantification of regeneration. Histo-electrophysiologic, and functional measures of

al.menthefroma stmunerternTheretu

Wfrauandtionrefldiscterncovesimrecoshoa nanedistcurrto atransciaticsion a

ry, 243 Charles Street, Boston, MA 02114. Fax: (801) 761--mail: thadlock@hotmail.com.

1926/01 $35.00t 2001 by Academic Presss of reproduction in any form reserved.CATION

ecovery Followingansection

. Hadlock1

and Harvard Medical School, Boston, Massachusetts

ctober 31, 2000

. Calculation of the SFI involves the measure-of various relationships between toes and feet of

dlimb of recovering animals. These are recordednked footprints left by the animals on the floor ofdardized walking track. Development of the for-s based on the observation that following sciaticinjury, rats develop characteristic walking pat-that can be reliably reproduced and measured.recovery can be followed by this gait analysis,ing to normal when full recovery is achieved.king track evaluations are cumbersome andt with technical problems in both performancealysis, and are therefore subject to error. Addi-y, the data obtained from SFI calculations are aion of complex integrated function, rather thante motor, sensory, or proprioceptive function. Al-ive methods for the assessment of functional re-following sciatic nerve manipulation, which arer, more quantitative, and distinguish type ofry, are therefore desirable. We have previouslythat following crush injury to the sciatic nerve,neurobehavioral test battery, developed in theesia literature, predicted functional recovery in-uishable from walking track analysis (8). In thet study, we sought to expand these observationsore extensive neural damage model, using a

ction injury.notch via a gluteal musculature splitting inci-nd sharply transected. Immediate microsurgical

epineurial repair was then performed, using two orthree 9-O nylon sutures. The muscles were reapproxi-mated and the wound was closed. During the survivalperiodBitterCT) toderwestandasor po36 we

Funassesstests wmal atbetweto ide(P , 0

Forbecom8.7 3Theirtion, aupon awere aBain eboth tNPL),both smiddleNIT).

SFI 5

All mblindeSomeby thecases,resultrecord

Themanceet al. (testingin a suThe rififth fithe animal wuponNJ) (sof 0 tois protpaper.weigh

denervated limb was recorded. The same was done forthe unoperated side. The formula for calculating thepercentage functional deficit is

rc

rest

ly.heiv

f-motEP

a windlifr

ghistalti

henrmllyemklekneriet

ryishet fk actreesyo

hesu

y te

ertssh30erorequ

censued

193BRIEF COMMUNICATION, autotomy was averted by the administration ofGreen taste deterrent (Grannicks, Greenwich,the toes of the affected hindlimb. Animals un-

nt periodic behavioral testing using both therd walking track analysis and the newer exten-

stural thrust (EPT) measurements, for a total ofeks.ctional recovery predicted by each method wased, and pair-wise comparisons using Students tere made among the two methods for each ani-each time point. Comparisons were also made

en baseline function and that following recovery,ntify the time at which statistically significant.05) recovery from baseline had occurred.SFI determinations, animals were allowed toe conditioned to the walking track, a wooden box43 cm in dimension and darkened at one end.hindfeet were dipped in a methylene blue solu-nd they were permitted to walk down the trackstrip of white paper. The prints left by the ink

llowed to dry and then analyzed as described byt al. (1). Measurements included print length onhe experimental and the normal sides (EPL,toe spread between the first and fifth digits onides (ETS, NTS), and the distance between theof the second and fourth toes on both sides (EIT,

The formula used to calculate SFI was as follows:

238.8~EPL 2 NPL!/NPL 1 109.5~ETS

2 NTS!/NTS 1 13.3~EIT 2 NIT!/NIT 2 8.8.

easurements were performed manually ind fashion and recorded to the nearest millimeter.prints were unmeasurable due to smearing of ink

tail, or contamination by the forefeet. In thesethe run was repeated to obtain interpretable

s. The maximum distances for each value wereed for each walking track.EPT was utilized as a reflection of motor perfor-and was measured as described by Thalhammer14). Specifically, each animal was placed on thesurface and the entire upper body was wrappedrgical towel. Only the hindlimbs were exposed.

ght hindlimb was supported with the examinersnger while the entire hand grasped the torso ofimal. With the left hindlimb suspended, the an-as held upright and the hindlimb was placed

a digital scale (Ohaus LS2000, Florham Park,ee Ref. 8, Fig. 1). Any digital scale with a range500 g is appropriate. The flat surface of the scaleected with a taped segment of heavy absorbentIt is clear when the animal begins to bear

t on the scale. The amount of weight borne by the

pe

whethrutive

Terat(sel

Manddatorighinicitweistattionany

TaroudeteticasurWeewhiweemaiinjubothcovemes

TmentracFundegmesonly

Tmeaeasbetwto pprintion20unda mit repermeaimment motor deficit 5 ~NEPT 2 EEPT!/NEPT,

NEPT and EEPT represent extensor posturalon the normal and experimental sides, respec-

re were no procedure-related deaths, no postop-e wound infections, and insignificant autonomyutilation) during the study period.

or recovery was analyzed by both walking trackT methods. Results are shown in Fig. 1. The SFIere plotted directly as calculated. The EPT data,

ally measured in grams of weight borne by eachmb (Table 1), were expressed as percentage def-om total bearing weight as determined by thet borne by the unoperated limb. There was noically significant difference in the level of func-recovery predicted by either testing modality atme point.re was a smaller average standard deviationd the EPT measurements as compared with SFI

inations. This led to earlier detection of statis-significant levels of recovery using EPT mea-

ents (Week 5 and beyond, with the exception of7, versus Week 12 and beyond for SFI). Overall,significant recovery was observed over the 36-study period, a substantial functional deficit re-d, as is expected following complete neurotmesiss. In our previous work using the crush model,

esting modalities had demonstrated complete re-over 4 to 6 weeks, as is expected with axonot-

injuries.current standard for functional recovery assess-ollowing sciatic nerve injuries has been walkingnalysis, which allows computation of the Sciatic

ion Index. It provides a quantitative measure ofof functional deficit, but can be cumbersome,

, time consuming, and variable. It also providesne integrated measure of function.EPT measurements also provide a quantitativerement of functional recovery, but are simple,o execute, and give consistently less variationen measurements. On average, the time it takesform walking track analysis, from recording the, cleaning the animal, and measuring the rela-ips, to calculating the SFI, totals approximatelymin per animal. The EPT testing is done in

30 s. Given its simplicity, it is easy to employ onfrequent basis. It has the additional benefit thatires minimal data manipulation to calculate thetage motor deficit. Without the cumbersomerements and calculation steps, it is easier toiately identify animals whose recovery differs

from tothermine rplex in

Potecertainfortabcritica

. Ta

sinll tninese

FIG. terblack ci s w

Rat N

Hindlim

Week2345678

1012141618202236

Note.

194 BRIEF COMMUNICATIONhat which is expected. It can be accompanied byneurobehavioral tests when desired, to deter-ecovery of nociceptive, proprioceptive, and com-tegrated function (3, 4, 9, 10).ntial pitfalls of the EPT do exist. It takes a

training period for the tester to become com-le handling the animals, and this comfort level isl for the animal to behave in a nonfrightened

wayanimcalat atraicom

Wtwe

1. Graph illustrating percentage functional deficit versus time, dercles with dashed error bars. EPT data are plotted in gray triangle

TABLE 1

o.: 1 2 3 4

b: L R L R L R L R

11 150 9 130 12 115 15 14013 192 15 192 16 134 20 1338 124 11 130 8 113 25 113

15 133 21 115 19 120 26 12218 119 20 120 12 110 14 11022 154 14 101 15 102 16 10613 95 21 110 15 102 15 12025 106 15 109 28 121 18 11737 113 33 106 27 110 29 10753 131 43 120 45 126 44 12040 135 27 120 55 141 41 11860 150 56 120 45 110 38 10573 150 58 131 45 110 42 9278 166 57 158 47 143 62 12286 120 72 130 29 116 68 140

Data represent weight borne (grams).here is also a level of recognition of when thel is bearing its maximum weight, which is criti-ce the tester is supporting the body of the animalimes. Once this recognition takes place, throughg by an experienced tester, the examination be-highly reproducible (3, 4, 9, 10, 14).had previously shown excellent correlation be-SFI and EPT determinations following rat sci-

mined by both SFI and EPT testing. SFI data are plotted inith solid error bars.

5 6 7 8

L R L R L R L R

12 116 13 111 13 113 7 9912 137 15 129 15 137 7 9512 116 14 102 18 122 8 10615 110 22 103 15 110 16 12016 115 20 94 10 118 11 1077 109 24 118 10 113 11 106

11 110 29 117 15 116 10 10411 110 20 117 18 121 12 12014 109 25 97 30 104 30 11321 129 45 127 33 110 35 15428 119 32 121 57 131 44 14330 115 37 118 55 135 47 14935 110 37 109 57 140 50 15035 137 47 141 75 138 57 14145 112 62 116 78 118 54 114en

atic nerve crush injuries. We have now demonstratedthat the same holds true for more extensive neuralinjuries. We conclude that there is a role for EPTmeasurements as a potential replacement for SFI mea-surements in the assessment of functional recoveryfollowing neural repair.

REFERENCES

1. Bain, J. R., S. E. Mackinnon, and D. A. Hunter. 1989. Func-tional evaluation of complete sciatic, peroneal, and posteriortibial nerve lesions in the rat. Plast. Reconst. Surg. 83: 129136.

2. Carlton, J. M., and N. Goldberg. 1986. Quantitative integratedmuscle function following reinnervation. Surg. Forum 37: 611.

3. Castillo, J., J. Curley, J. Hotz, M. Uezono, J. Tigner, M. Chasin,R. Wilder, R. Langer, and C. Berde. 1996. Glucocorticoids pro-long rat sciatic nerve blockade in vivo from bupinvacaine mi-crospheres. Anesthesiology 85: 11571166.

4. Curley, J., J. Castillo, J. Hotz, M. Uezono, S. Hernandez, J. O.Lim, J. Tigner, M. Chasin, R. Langer, and C. Berde. 1996.Prolonged regional nerve blockade. Injectable biodegradablebupinvacaine/polyester microspheres. Anesthesiology 84: 14011410.

5. Dellon, A. L., and S. E. Mackinnon. 1989. Selection of theappropriate parameter to measure neural regeneration. Ann.Plast. Surg. 23: 197202.

6. DeMedinaceli, L., W. J. Freed, and R. J. Wyatt. 1982. An indexof the functional condition of rat sciatic nerve based on mea-surements made from walking tracks. Exp. Neurol. 77: 634.

7. Ellis, J. C., and T. V. McCaffrey. 1984. Animal model for pe-ripheral nerve grafting. Otol. Head Neck Surg. 92: 546.

8. Hadlock, T. A., R. Koka, J. P. Vacanti, and M. L. Cheney. 1999.A comparison of assessments of functional recovery in the rat.J. Periph. Nerv. Syst. 4: 258264.

9. Hu, D., R. Hu, and C. B. Berde. 1997. Neurologic evaluation ofinfant and adult rats before and after sciatic nerve blockade.Anesthesiology 86: 957965.

10. Lee, K. C., R. T. Wilder, R. L. Smith, and C. B. Berde. 1994.Thermal hyperalgesia accelerates and MK-801 prevents thedevelopment of tachyphylaxis to rat sciatic nerve blockade.Anesthesiology 81: 12841293.

11. Mackinnon, S. E., A. R. Hudson, R. E. Falk, and D. A. Hunter.1982. The nerve regeneration model and trophic factors in vivo.Brain Res. 13: 334.

12. Mackinnon, S. E., A. R. Hudson, and D. A. Hunter. 1985.Histologic assessment of nerve regeneration in the rat. Plast.Reconstr. Surg. 75: 384.

13. Munro, C. A., J. P. Szalai, S. E. Mackinnon, and R. Midha. 1998.Lack of association between outcome measures of nerve regen-eration. Muscle Nerve 21: 10951097.

14. Thalhammer, J. G., M. Vladimirova, B. Bershadsky, and G. R.Strichartz. 1995. Neurologic evaluation of the rat during sciaticnerve block with lidocane. Anesthesiology 82: 10131025.

195BRIEF COMMUNICATION

FIG. 1TABLE 1REFERENCES