Extract
Postoperative nerve palsy associated with revision total hip arthroplasty is a serious neurologic complication, occurring in up to 50% of cases. While a variety of intraoperative monitoring techniques have been used to reduce the prevalence of this complication, they are not sensitive or accurate enough to prevent surgical damage to the sciatic nerve. A new, modified intraoperative montage utilizing spontaneously elicited electromyography (sEMG) and near-nerve action potentials (NAPs) was evaluated in a retrospective review of thirty-five consecutive revision total hip arthroplasties.
Postoperative nerve palsy associated with revision total hip arthroplasty is a serious neurologic complication, occurring in up to 50% of cases. While a variety of intraoperative monitoring techniques have been used to reduce the prevalence of this complication, they are not sensitive or accurate enough to prevent surgical damage to the sciatic nerve. A new, modified intraoperative montage utilizing spontaneously elicited electromyography (sEMG) and near-nerve action potentials (NAPs) was evaluated in a retrospective review of thirty-five consecutive revision total hip arthroplasties.
SEMGs were recorded from sterile needle electrodes placed in the tibialis anterior, extensor hallucis longus, and gastrocnemius muscles of the involved extremity. NAPs were obtained from needle electrodes placed near the peroneal nerve (knee) and tibial nerve (ankle). Audible recordings were monitored, and the surgeon was informed of the location and stimulus of physiologically important occurrences.
Four incidents of sEMG firing and one episode of NAP occurred during surgical exposure. The surgical approach was redirected, and all firings subsided spontaneously. Seven episodes of sEMG firing occurred during retraction. Two resolved spontaneously, and five resolved with repositioning of the retractor. Two episodes of sEMG firing occurred during lengthening; both resolved with reduced lengthening. Five episodes of sEMG firing occurred during wound closure. All five resolved with loosening of tissue tension.
Fourteen instances of potentially irreversible nerve damage were recorded during this study. With appropriate warning from this nerve-monitoring protocol, action was taken such that no neurologic deficit occurred in any of the thirty-five patients who were monitored. The next twenty-eight sequential patients were reviewed prospectively to increase the sample size and to evaluate the learning curve.
The number of orthopaedic surgical procedures in which the sciatic nerve is manipulated is increasing rapidly. The nerve is at risk not only during revision total hip replacement but also during many operations done for hip dysplasia and for pelvic and acetabular fractures. The prevalence of minor nerve injury during revision hip surgery has approached 50% to 70% in some studies 1,2 . Many of these injuries are transient stretch neuropraxias, but each irreversible event can give rise to a lawsuit. While a variety of intraoperative nerve-monitoring techniques have been used to reduce the prevalence of injury, they are not sensitive or accurate enough to prevent surgical damage to the sciatic nerve. Somatosensory evoked potentials (SSEP) can be affected by anesthesia, spinal/epidural blocks, and even room temperature 3,4 . EMG alone may miss some proximal nerve events if the specific muscle involved is not among those sampled. A new, modified intraoperative montage utilizing sEMG and NAP was evaluated in a retrospective review of thirty-five consecutive revision total hip arthroplasties. An additional twenty-eight patients were evaluated prospectively, with use of the same protocol in a more efficient manner, for a total of sixty-three consecutive revision total hip arthroplasties.
All sixty-three patients gave informed consent to undergo intraoperative nerve-monitoring with wire leads on the involved side during revision hip surgery. The needle electrodes that were used were typical half-inch (1.3-cm) EMG leads. The involved leg and foot were fully prepared and draped, and the leads were then inserted over the peroneal nerve and tibial nerve (for the monitoring of NAPs) and over the anterior tibialis, extensor hallucis longus, and gastrocnemius muscles (for the monitoring of sEMGs), to cover the major divisions of the sciatic nerve ( Fig. 1 ). The electrical input was displayed with use of four channels of a five-channel TECA monitoring computer (Pleasantville, New York) ( Figs. 2 and 3 ). Waveforms were monitored visually and EMG signals were monitored audibly by the technician and the surgeon. An abnormal signal was immediately reported to the surgeon, and changes in positioning, retraction, or other nerve-irritating factors were made until the signal returned to baseline ( Figs. 4 and 5 ).
In the first thirty-five patients, four incidents of sEMG firing and one episode of NAP occurred during surgical exposure. The surgical approach was redirected, and all firings subsided spontaneously. Seven episodes of sEMG firing occurred during retraction. Two resolved spontaneously, and five resolved with repositioning of the retractor. Two episodes of sEMG firing occurred during lengthening. Both resolved with selection of a shorter head-neck component. Five episodes of sEMG firing occurred during closure. All five resolved with loosening of tissue tension.
In the next twenty-eight patients, one episode of sEMG firing and two episodes of NAP occurred during retraction, one episode of sEMG firing and one episode of NAP occurred during reduction, and one episode of sEMG firing occurred during closure. The timing of all twenty-five episodes that occurred during the sixty-three procedures is given in Table I .
One patient had a permanent femoral nerve palsy following an anterior approach. This nerve had not been monitored intraoperatively. One patient had a footdrop two days postoperatively. This complication was due to an excessively tight abduction pillow strap and completely resolved by the three-month examination and EMG. One patient had a 65% reduction in the preoperatively monitored baseline motor unit potentials and decreased amplitude in the right near-nerve response, consistent with the preoperative diagnosis of diabetic neuropathy. These were unchanged postoperatively.
The average duration of surgery decreased by fifty minutes from the first group to the second group ( Fig. 6 ).
Sciatic nerve-monitoring with this protocol provided instant indications of potential permanent nerve damage during 30% (nineteen) of the sixty-three revision total hip arthroplasties that were studied. The majority (60%) of events occurred during exposure and retraction. Eighty-four percent of the potential injuries were related to exposure and closure and not to reduction and leg-lengthening. In our experience, monitored neurophysiologic events rarely occur once complete exposure has been obtained, unless the deep posterior retractors are moved. Therefore, the most critical times to monitor are the approach and exposure portions of the operation.
Initial placement and connection of the monitor leads added twelve to fifteen minutes of time to each procedure. With repetition, the process has been streamlined to less than five minutes.
Although the two study groups were not matched for severity, the average surgical time between the two groups decreased by 30% (p < 0.001). There was a definite increase in the speed of obtaining surgical exposure in the second group as confidence grew in the monitoring technique, and we believe that the monitoring can realistically decrease surgical time by twenty to thirty minutes.
The limitation of current recording protocols focuses primarily on the number of recording channels available to assess electrophysiologic data. This limit of input channels created an inability to monitor and note permanent femoral nerve stretch and damage in one patient in our study. Most current protocols can only utilize two to five TECA channels for data recording.
In an attempt to optimize intraoperative recording during revision total hip arthroplasty, we have developed an idealized recording system capable of inputing eight channels of electrophysiologic (audio waveform) data ( Figs. 7 and 8 ). Our data are currently analyzed on the basis of accumulated intraoperative experience to produce a rules-based, audible, early warning to the surgeon when signals of neurophysiologic consequence occur, and our system will archive the intraoperative data to digital video media. We envision this technique being applied also to cervical and lumbar pedicle screw placement, acetabular fracture repair, and deep shoulder dissection.
Twenty-five episodes of potentially reversible sciatic nerve injuries occurred during nineteen (30%) of sixty-three monitored revision total hip arthroplasties. With appropriate warning from this nerve-monitoring protocol, action was taken such that no sciatic neurologic deficit occurred in any of the sixty-three patients monitored. From this experience, we conclude the following.
1. An intraoperative sEMG and NAP montage provides very effective, immediate evidence of potential nerve injury.
2. This method avoids the need for cortical stimulus (SSEP) and eliminates the negative effects of anesthesia and nerve blocks on distal sciatic monitoring.
3. If monitoring is not available, extra care should be utilized during the exposure and closure stages of revision total hip arthroplasty.
4. With use of this monitoring technique, surgical exposure time decreases by about twenty minutes.
5. There is a learning curve for the surgical team to become confident in the accuracy of the monitoring technique.
6. Five or six channels are needed to completely monitor the involved extremity.
7. The risk of sciatic nerve injury during revision total hip arthroplasty can be significantly reduced with this method of intraoperative neurophysiologic monitoring.
DeHart MM,Riley LH Jr. Nerve injuries in total hip arthroplasty. J Am Acad Orthop Surg,1997;7: 101-11. 7101
1997
Kennedy WF, Byrne TF, Majid HA,Pavlak LL. Sciatic nerve monitoring during revision total hip arthroplasty. Clin Orthop,1991;264: 223-7. 264223
1991
[PubMed]
Helfet DL, Anand N, Malkani AL, Heise C, Quinn TJ, Green DS,Burga S. Intraoperative monitoring of motor pathways during operative fixation of acute acetabular fractures. J Orthop Trauma,1997;11: 2-6. 112
1997
[PubMed][CrossRef]
Sutherland CJ, Miller DH,Owen JH. Use of spontaneous electromyography during revision and complex total hip arthroplasty. J Arthroplasty,1996;11: 206-9. 11206
1996
[PubMed][CrossRef]