Intraoperative cortical somatosensory evoked potentials for detection of sciatic neuropathy during total hip arthroplasty

3eparement of AnestResiology and the Section 0: Orthopedic Seargery, University of Kansas Scboo of Medicine, Kaxsas City, KS.

3 $70ofpatients undergoing hip ar~~~op~~t9 develop postoperative sciatic neuropathy. The factors associated with ~~~~ges in sQmat~se~s~~ evoked ~~te~t~a~ (SSEP) and sciatic ne~r~~at~~ were examined in patients ~~de~g~~~g Approximute~9

hi$ arthroplasty~ to evaluate whet ep_the we of ~~traoperat~~e SSEP could help reduce the incidence of postoperative sciatic ne~ropath~. ~~g~ty-e~g~t patients were *Assistant

Professor

tllesident iProfessor

and Chairman

PWesearch Assistant Address reprint requests to Dr. Porter at ihe Department of Anesthesiology, University of Kansas School of Medicine, 39th and Rainbw Boulevard, Kansas City, KS 66103, USA.

assigned to either monitored or unmonitored groups. SSEP were re~~~dedfo~~~wi~~ peroneak nerve st~~~~ati5n, using contralateral stimulation to detect systemic injluences on SSEP. Amplitude reduction of GO% of control andlor latency increase of >PO% of control was considered si~if~ca~t, and su caE ~~te~e~ti~~ was attempted to restore SSEP. Previous surgery and a lateral incision a~~r#a~~ tended thy (p < 0.053). The incidence of.sciatic to be associated with sciatic ne 3 %,,Iwas not different from the u~,~o~itored ne~~~~at~~in the monitored g~o g-rou$~(2.4%). Isolated reduction in amplitude or ~ro~~~gat~o~ in Eatency of the SSEP was not predictive of ~~stoperat~tieneurologic ,function of the sciatic nerve. Six patients, two of whom developed sciatic ~e~~opa~~~, demonstrated ~~m~~ete flattening

of the SSEP. Both of these patients had~attened

SSE

surgical events (p < 0.01) and ~attev~ed SSEP were present at the end of the surgical procedure. There were no false-negative SSEP changes. S~r~~~tane~~ amplitude and latency changes appear to be predictive of sciatic nerve ,f~n~~io~

Received for publication August 27, 1988; revised manuscript accepted for publication November 4, 1988. 0 1989 Butterworth

J. Clin.

Publishers

Anesth.;

1989,

vol. 1, no. 3

following

hip arthro~l~ty.

is investigation

was

PQ$toFerative neuropath plication of rotd h nce

of injury

ing this procedure range from 0. on clinical syndrome involves the prognosis for recovery is generally t 75% of patients full ies of 2,012 patients, erative palsies, 7 sciatic, 3 femoral, mixed femoral/sciatic. All but one sod function. In a subs r et ai.’ demonstrat electromyographic evidence of throplasty, although none s. In 540 cases performed I984 and 1986, the inciic nerve palsy was 2.6%. y was found to be more common in 5) and in those undergoing repeat ame hip (p < 0.05)’ nisms of sciatic nerve injury have the sciatic nerve by neural blood flsw, as res varying from 50 to 80 Brenner observed that leion of the sciatic nerve roneal division,‘O and micronerve demonstrate reduced s enhancing its vulnerability injury is likely to occur is ~~itialIy dislocated, when retractors acetabular and femoral prepara6 the hip, especially if the ive dislocation, intraneural bleeding in anticoagulated patients, beat generated by methy~m~t~ac~late~ and extrusion of cement near or around the nerve can also be implicated in some cases3,i2J3 tosensory evoked responses (SSEP) traoperative neural function began over a decade ago, and is widely utilized for assessment of cortical, spinal, and p 1nerve function during a variety of operative res in which the function of the central or pe were first utilized during et &.,I5 who reported sigve SSEP changes in I2 of 50 patients All SSE had resolved by the en no patient develop ostoperative nerve injury. ore specific postoperative ~~~r~~athies bave not been subsequently ~~e~ti~e~~

ative sciatic nerve injury?

After iustit~t~oua~ approval were obtained, 88 adult AS

intubation

was facilitated

ere monitored in all SSEP were recorded

common manifestation

of s

ses were recor

threshold.

delivered ranstimuli oer second at three Ban were set at 18

ocedures were

i-y. A baseline response was recorded after each quenl responses were recorded after dislocation of

Original Contribuliom

from the operated extremity was thought “0 be significantly altered, the surgeon was notifie an attempted to intervene to reesta normal SSEP morphology. Intervention include ,oosition, retractors, or the prosthesis.

patient

stepwise discriminant ciation between the presence of sciatic neuropathy was examined exact probability test.

PLRONEAL [STIMULATOR)

igure B. Montage for recording SSEP during total hip arthroplasty. Stimulating electrodes were placed over the peroneal nerve at the fibular neck bilaterally.

t ip, which may stretch the sciatic nerve. were next recorded during acetabular prep ration and reaming, during which time compressed posterior to the Holman nses were repeated after relocation of the hip, which may stretch the nerve further, especially if the leg has been lengthe were last recorded during skin closure. stimulation of the contralateral peroneal nerve were used for detection of systemic influences on SS Latency of the N 1 cortical response and amplitude of the N 1Pl component of the cortical response were tabulated for each event, Reduction in amplitude sf over 50% of baseline amplitude and increase in latency of over B0% of baseline latency were considered significant changes in SSEP that might indicate comnromise in sciatic nerve function.15 When the SSEP I.72

J. Clin. Anesth., 1989, vol. 1, no. 3

grcpizp were

ana-

be assonse and

of prosthesis used (Table d itored group were diagn ving avascular necrosis (p < 0.05). One patient in the un niitored group sustaine a permanent sciatic nerv Lsy (2.3%). Two patients (4.6%) in the monitored group awoke with motor weakness indicative of sciatic nerve injury (foot and both recovered some degree of functio difference in the incidence of sciatic neuro hween the two groups was not statistically si There was no significant association isolated patient characteristic or surgica the occurrence of sciatic nerve injury, a’ktbou patients who sustain previous surgery on had been approached using a lateral incision and exposure. h SSEP were recorde from both CZ and G7, 67 responses were tee nically i~ade~~~te in Over 50% of the patients due to motion artifact. Gortical responses w-ere therefore selected for analysis. Almost 50% of the monitored patients showed significant changes of cortical SSEP at some tune during the procedure. There were 56 significant changes in EP measurements (Table SSEP out of a total of 46 ) were changes in

on the nonoperated

extremity

co

red to baseline

Characteristics

of Monitored

and Unmonitored

(n = 46)

(n = 42)

18 (39) 28 (61)

21 (50) 21 (50)

(mean + SEM)

56.4 k 2.6

59.5 k 2.7

Weight (kg) (mean ” SEM)

72.4 ?I 2.1

78.0 + 2.8

22 (48) 24 (52)

24 (57) 18 (43)

15 (33) 31 (67)

13 (31) 29 (69)

13 22 10 1

3 22 16 1

Sex Male FernaPe

Age ON

Side Right Left

Posterolateral Diagnosis AVN

DJD

Failed TW Fracture

(28)X (48) (22) (2)

(7) (52) (38) (2)

~~rn~~tan~Q~§ significant amplitude reduction and rolongation of the SSE response at some ng the surgical procedure (15% of all SSEP changes) in six patient, this change manifested a of the response (Figure 2). These c es were not found to be significantly associated event and were not predictabl or surgical characteristics in t patients demonstrated recovdex intervention. The two remaining Cents awoke with sciatic nerve injury. These tw subjects

Proslhesis type Cernentiess Hybrid

26 (57) 20 (43)

22 (52) 20 (48)

previous surgery Yes No

22 (48) 24 (52)

18 (43) 24 (57)

2 (4.3)

1 (2.4)

Postoperative

Isolated abnormalities sure in 10 patients des vention. None of these

Significant SSEP change intraoperatively

flattening

for two or more

was a highly significant a act probability test (p < Fortant abnormalities o in closure; one response e other recovered to larency remained signi

sciatic

palsy

who demonstrated

that persisted

of the response

surgical events. This

con&u

NIA

21 (46)

AYN = avascular necrosis; DJD = degenerativejoint disease; THR = total hip replacement; N/A = SSEP not monitored. *$J < 0.05; compared to unmonitored group.

Number of Significant Dislocation

The development of si tency abnormalities in

Changes in SSEP (9% of aI1 SSEP changes)

Acetabular

Amplitude (<50% of controi) Latency (>lO% of control) Amplitude (~50%) and latency (>lO%)

J. Clin. Anes:h.,

1989, veL j) no. 3

17

Chgarml Contribution4

tion between

specific quantrtatnve ~~~~~~~~~~i~~~ changes e peripheral nerve in.u po transient changes in bo !.atency of cortical S operative monitoring, serious ab~~rrna~~ti~ tical

Figure 2. This patient demonstrated flattening of the SSEP during reduction of the hip, with recovery during skin ciosure. The patient had no postoperative neurologic deficit.

10% control) foollow oneal nerve stimulation was highly predictive o operative sciatic nerve dysfunction, in that all who developed flattened SSEP for two or more surgical events awoke with sciatic neuropathy. In a Con, although 10 patients demonstrated isolated S abnormalities that persisted during skin closure, only the two patients who developed sciatic neuropathy exhibited serious abnormalities ofboth amplitude and latency at the time of skin closure. It is believed that his finding is the initial demonstration of a correla-

cedure in most series have correlated well with operative motor and sensory function. Allen & reported 11 of 12 patients with transient SSEP changes during surgery, and one in whom the cortical SSE (elicited by peroneal nerve stimulation tion of a spinal cord tumor) was lost recover. This patient demonstrated postoperative mocror weakness of the monitored exerern~~y~ Grundy stated that “irreversible obliteration of SSEP has invariably been associated with a postoperative neurologic deficit in the monitored pathway.“17 She reported that one patient was left paraplegic after attempter resection of a spinal arterial venous malformation g which the cortical S audzens described mul transient latency and a itude changes unrelated to surgical manipulation.rg ne patient with irreversible loss of SSEP compon neurologic deficit. Whittle et aE.*Onoted that onl sistent loss of waveform components with an acquired neurologic deficit, w tions in amplitude of up to 7’0% of the basehne value agnostic importance. Our results are supthese prior results, confirming the nostic value of simultaneous irreversible att~~~ati~~ of amplitude and latency pro~o~~at~on.

Figure 3. Intraoperative SSEP from the two patients who developed postoperative sciatic neuropathy. (A) SSEP was initially flattened during acetabular preparation and remained so until reduction of the hip. Although there was partial recovery during skin closure, significant amphtude rednction and latency prolongation persisted. (B) SSEP was attened during hip reduction and skin closure.

134

j. Clin. Anesth.,

1989, vol. 1, no. 3

cm-

§SEP

thers’“~‘” have emp asized the importance of utind cmtrols during SSE eliminate systemic, metabolic, and electrithat influence SSEP. Only by monitoring y the surgical procedure can such influences as blood pressure, oxygenation, temperature, and a~e§thet~c effects be identified. This ortant because surgical interrule was especially i ventions were to be de based on changes in SSEP morphology. onitoring reflects only the integrity of tbe a&ways, in this case the p sal coiumns of the spinal co tosensory cortex. There is both experimental and clinence that sensory and motor changes due to or compression of spinal cord correspond closely.23,24 Nevertheless, there are isolated reports of neur c deficits in motor function tha pred traoperative SSEP,25,26 and in 167 cases ed that 2.9% ofrecordings ive.27 Cortical SSEP changes may more eripherai motor function during otor and sensory fibers are ip artbsoplasty because arate in the periphera! nerve, as e spinal cord. In the present study, there P changes in monitoring ty. Because intervention of SSEP abnormality to ossibility of neuropathy, the true incidence tive or fake-negative changes cannot be sciatic neuropathy is an inip arthroplasty, it is potene. The ability to detect nerve e reversible is the basis for the use of intraoperative SSEP. This study was unable to demonstrate a reduction in the incidence of sciatic zing evoked potential monitoring. emonstrates that well over 3,000 paave to be studied

to detect

a reduction

ciatic palsy from 3.0% to 0. I % (Table monitor patients undergoing hip

le 3. Number of atiens Necessary to Detect a eduction in Postoperative Sciatic Nerve Injury No.

3.Q to 2.0 3.0 to 1.0 3.0 to 0.1

of Patients

Power 0.90

Power 0.80

42,837 10,710 5,094

30,926 7,732 3,678

Iasty will not only ai of ind negative rate and t e vn~~era~~~~~~l Cents to sciatic nerve injury, but will also the ultitnate utility of evoked potentHa Isolated changes in either a refore be considese r studies should evaluate the reliability multaneous amplitude an these results, a wbicb, fro dictive value for posts

statistical analysis and preparation

of si-

D, for assistance in of the manuscript.

R, Coventry MB. Peripheral neuropathies associated with total hip arthroplasty. J Bone joint Surg ~976;5ak66-9. Collins WRY Nef2. BeHaven KE, Evarts CM, Wilde A ~har~~ey-~u~~er son 6, Razzano CD: Early results total hip reconstruction. Ortkop Cl& North Am 1373;

_. Weber EW, Daube

4:4X-72. 3. Amstutz HC, Ma al L: evision of ies. in Orthop aseptic loose tot 1982;170:21-33. 4. Solheim LF, Hagen : Femoral and sciatic neuropathies after total hip arthroplasty. Acka Ortkop &and 1980;51:531-4. .s. Eftekhar NS, Stinchfield FE: Experience with low-friction arthroplasty. A statistical review of early results and complications. Clin Orthof 1973;95:58-8. ten year results of 6. Charnley J, Cupic Z: The nine a _e hip. Glin Whop the low friction arthroplasty of L973;95:9-25. 7. Bruce P, Bramble JM? Ret g EW: Nerve palsies associated with total hip arthr lasty. Presented at the Meeting, Jan 22-27, Am Acad Orthop Surg Ann !987. 8. tundborg G, Rydeuik I%: Effects of stretching the tibia1 nerve of the rabbit. A preliminary study of intraneural circulation and the bar ion ofthe perineurium. -401. ,jrBonejoint Surg 1973 9. Rydeuik B, Lundborg ogge 8;: Effects of graded compression on intraneuraf bloo flow. J Wand Surg i981;6:3-12. IO. Denny-Brown D, Brenner C: ParaEysis of nerve induced by direct pressure and by tourniquet. Arch, Neural Psyc;kiatry 1944;51: l-26. I?. ~Sunderfand S: Blood supply of the sciatic nerve and its popliteal divisions in man. Ad Neural Psychiatry !945;54:283-9. _I. Clin. Anesth.,

1983, 1701. i, no. 3

175

Contributions

Original

12. Eazansky MG: Complications revisited. The debit side of total hip replacement. Clin Srlhop 1973;95:96-103. 13. Fleming RE, Michelsen CB, Stinchfield FE: Sciatic paralysis. A complication of bleeding following hip surgery.g Bone Joint Surg 1979;61A:37-9. 14. Grundy BL: Intraoperative monitoring of sensory evoked potentials. Anesthesiology 1983;58:72-87. 15. Stone RG, Weeks LE, Hadju M, Stinchfield FE: Evaluation of sciatic nerve compromise during total hip arthroplasty. Clin Qrthop 1985;201:26-3 1. 16. Allen A, Starr A, IVudleman K: Assessment of sensory function in the operating room utilizing cerebral evoked potentials: A study of fifty-six surgically anesthetized patients. Clin Neurosurg 1981;28:457-81. 17. Grundy BL: Evoked potential monitoring. In: Mitt CD, ed. Monitoring in anesthesia and critical care medicine. New York: Churchill-Livingstone, 1985:384-8. 18. Grundy BL, Nelson PB, Doyle E, Procopio PT: Intraoperative ioss of somatosensory evoked potentials predicts loss of spinal cord function. Anesthesiology 1982; 57:321-2. 19. Raudzens PA: Intraoperative monitoring of evoked potentials. Ann NY Acad Sci 1982;388:308-26. 20. Whittle IR, Johnstone III, Besser M: Recording of spinal somatosensory evoked potentials for intraoperative spinal cord monitoring. J Neurosurg 1986;64:601-12.

176

J’

J Clin. Anesth.,

1989, vol. I, no. 3

21. Kaplan BJ, Freidman WA, Alexander JAY !-lampson SW: Somatosensory evoked potential monitoring during repair of aortic coarctations. Proceedings of Evoke Potential Monitoring: State of the Art 1985. Sept 1821, 1985, Lake Buena Vista, FL. 22. Young W: False-negative results for intraoperative SSEPs (Letter to the Editor).J Neurosurg 1986;64:987-8. 23. Cheng ME, Robertson 6, Grossman RG, et al: Neurologic outcome correlated with spinal evoked potentials in a spinal cord &hernia model. J Neurosurg 1984;60:786-95. 24. Golles JG, Wilson GJ, Sims AF, et al: Intraoperative detection of spinal cord &hernia using §orn~~o~e~§ory cortical evoked potentials during thoracic aorta occlusion. Ann Thorac Surg 1982;34:299-306. 25. Ginsburg HH, Shetter AC, Raudzens PA: Postoperative paraplegia with preserved intraoperative somatosensory ev otentials. J Neurosurg I985;63:2~6-3~~. 26. Lesser audzens P, Luders II, et al: Postoperative neurological deficits may occur despite unchanged inrraoperative somatosensory evoked potentials. Ann Neural 1986;19:22-5. 27. Hahn JF, Latchaw JP: Evoked potentials in the operating room. Clin Neurosurg 1984;3 1:389-403.