Edited by: Jaimo Ahn, Hospital of the University of Pennsylvania, USA
Reviewed by: Konstantinos Markatos, Henry Dunant Hospital, Greece; Harish Hosalkar, The Hosalkar Institute for Joint Preservation and Injury Care, USA
Specialty section: This article was submitted to Orthopedic Surgery, a section of the journal Frontiers in Surgery
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Intraoperative neuromonitoring (IONM) has become a standard of care in spinal deformity surgeries to minimize the incidence of new onset neurological deficit. Stagnara wake up test and ankle clonus test are the oldest techniques described for spinal cord monitoring, but they cannot be solely relied upon as a neuromonitoring modality. Somatosensory evoked potentials monitor only dorsal tracts and give high false positive and negative alerts. Transcranial motor evoked potentials (TcMEPs) monitor the more useful motor pathways. The purpose of our study was to report the safety, efficacy, limitations of TcMEPs in spine deformity surgeries, and the role of a checklist.
Retrospective review of all spinal deformity surgeries performed with TcMEPs from 2011 to 2015.
All patients were subjected to IONM by TcMEPs during the spinal deformity surgery. Patients were included in the study only if complete operative reports and neuromonitoring data and postoperative neurological data were available for review. An alert was defined as 80% or more decrement in the motor evoked potential amplitude, or increase in threshold of 100 V or more from baseline. The systemic and surgical causes of IONM alerts and the postoperative neurological status were recorded.
In total, 61 patients underwent surgery for spinal deformities with TcMEPs. The average age was 12.6 years (6–36 years) and male:female ratio was 1:1.3. Diagnoses included idiopathic scoliosis (
IONM alerts are frequent during spinal deformity surgery. In our study, more than 50% of the alerts were associated with anesthetic management. IONM with TcMEPs is a safe and effective monitoring technique and wake up test still remains a valuable tool in cases of a persistent alert.
Neurological deficit following surgical correction of deformity is a major concern for any spine surgeon (
After approval from Ethics committee (ID: EC/01/17/1107), retrospective review of all spine deformity surgeries performed in our institute during the period 2011–2015 was done. Our study included 67 deformity correction surgeries performed by three senior spine surgeons with a minimum experience of 15 years. Surgeries performed with TcMEP monitoring alone are included in our study. All the surgeries were performed under total intravenous anesthesia (TIVA) protocol developed by the institute, and a trained neurophysiologist who monitors IONM with TcMEPs. Age at the time of surgery, gender, diagnosis, duration of surgery, preoperative neurology, type of instrumentation, blood loss and the number of alerts during surgery, nature of insult, corrective measures done, and postoperative neurology were reviewed. From anesthesia records, depth of anesthesia and mean arterial pressure (MAP) at the time of alert plus anesthesia drug bolus usage were noted.
Before induction, the Stagnara wake up test is explained to each patient; TIVA was employed for induction and maintenance in all the patients. Anesthesia is induced with propofol 1–2 mg/kg i.v., fentanyl 2–3 μg/kg i.v., and dexmedetomidine 1 μg/kg i.v. Intubation is facilitated with only a small, single, short-acting dose of muscle relaxant. The patient’s eyes are taped shut and padded for protection from injury in the prone position. A urinary catheter is placed, an arterial line inserted, two large bore i.v. lines are secured, a temperature probe inserted, and appropriate sized bite blocks are wedged in place between the molars to prevent injury to the contents of the oral cavity (the teeth, tongue, and endotracheal tube). Intraoperative depth of anesthesia was judged by the bispectral index. All used sponges were weighed and saline washes measured, so that accurate assessment of intraoperative blood loss is made. Arterial blood gas analysis and hemoglobin estimations are done as and when required. Anesthesia maintenance is done with i.v. propofol 100–150 μg/kg/h, fentanyl 1–2 μg/kg/h i.v., and dexmedetomidine 0.5 μg/kg/h i.v.
Potentials were elicited by transcranial stimulation using corkscrew electrodes placed subcutaneously over the motor cortex (Nim-Eclipse, Medtronic). Motor evoked potentials (MEPs) were obtained from intramuscular electrodes (13 mm, 27G, dual electrodes) placed in four (sometimes five) bilateral muscle groups. One muscle group above the level of surgery was always used as a control (thenar muscles). Other electrodes were placed in rectus abdominis, vastus lateralis, tibialis anterior, and abductor hallucis. The most distal electrodes were placed in the anal sphincter in one case with S2 hemivertebrae. Ultrasound guided placement of electrodes into the rectus abdominis muscle was done in six patients.
Biphasic stimuli were given starting at three pulse, 200 V, and 0.5 ms duration with 2.0 ms interval between stimuli, and if needed increments were done each time by 25 V (up to 400 V) and at five- or seven-pulse train till a satisfactory baseline amplitude (50 μV) was obtained. The same protocol was followed during intraoperative monitoring and the maximal stimulus intensity needed was noted. The initiation of MEP stimulation and recording was done after intubation while the patient was in supine position and once again after patient was placed prone. The MEP recording obtained just before incision was taken as the baseline for future reference. The final MEP was obtained after the closure of wound but before application of dressing.
An “alert” was defined as a decrease in amplitude by 80% or more, or 100 V increase in threshold, or latency prolongation >10% from baseline in one or more electrodes. This need not necessarily be due to a surgical maneuver.
Parallel alert: a similar change (increase/decrease) seen in all the recording electrodes.
Non-parallel alert: a decrease or loss seen in only one or few recording electrodes.
There is an ongoing protocol in the hospital as a part of neuromonitoring program in Department of Spine Surgery that defined these alerts and also a protocol taken in response to an alert (Figure
When an alert was noted immediately following a high-risk maneuver and if recovery of amplitudes was noted after undoing that maneuver, rest of the parameters in the checklist being normal, then it was considered to be the cause. However, if an alert was noted during a routine monitoring protocol, respective teams evaluated all the parameters and the corrective maneuver by which the amplitudes were restored was taken as the most probable cause of an alert.
If the alert persisted even after all the corrective measures were undertaken for up to 30 min, the Stagnara wake up test was done. If the test was negative, the surgery was continued while MEPs were obtained at regular short intervals and if the wake up test was positive, surgery was aborted and the attendants were explained regarding the same.
The success of IONM (TcMEPs in our study) in determining cord compression at an early stage is expressed with true positive (TP), true negative (TN), false positive (FP), and negative (FN).
TP: an alert that persisted despite corrective measures or returned to baseline after corrective measures, but patient had a positive wake up test (if performed) or postoperative new neurological deficit.
FP: an alert that persisted during surgery despite corrective measures, but patient had a negative wake up test (if performed) or developed no new postoperative deficit.
TN: no alert was recorded during surgery and patient developed no new neurological deficit following surgery.
FN: no alert was recorded during surgery, but patient developed neurological deficit following surgery.
Indeterminate: An alert that returned to baseline value following corrective measures and patient had no new postoperative neurological deficit.
Specificity (Sp), sensitivity (Sn), negative predictive value (NPV), and positive predictive value (PPV) were calculated in our study. Sp and Sn give the percentage of negative and positive outcomes correctly indicated by the technique. PPV and NPV describe the probability that a patient has an injury if the test is positive and does not if the test is negative, respectively. PPV and NPV describe the performance of the technique (chance of a positive or negative neurological event).
Safety was evaluated by observation for scalp burns, arrhythmias, or injuries due to movements induced by TcMEPs like tongue or lip lacerations, seizures, and whether these movements interfered with surgery.
A total of 67 patients underwent deformity correction surgery with TcMEP monitoring, 6 patients had preoperative neurological deficit and were excluded. A total of 61 patients are included in this study, with an average age of 12.8 years (6–36 years). Most common cause of deformity was idiopathic scoliosis (
Characteristics of patient population | |
---|---|
Age (years) | 12.8 |
Male:female | 1:1.3 |
Diagnosis | No. of patients |
Idiopathic scoliosis | 35 |
Congenital scoliosis | 13 |
Congenital kyphosis | 7 |
Congenital kyphoscoliosis | 4 |
Post-infectious kyphosis | 1 |
Post-traumatic kyphosis | 1 |
Minimal stimulus intensity required for baseline potential |
||
---|---|---|
Number ( |
Percentage | |
200 V | 8 | 13 |
250 V | 10 | 17 |
275 V | 16 | 26 |
300 V | 22 | 36 |
350 V | 5 | 8 |
We had a total of 33 alerts in 22 patients (36%) (Table
Inciting events for TcMEP alert |
||
---|---|---|
Number ( |
Percentage | |
Hypotension | 7 | 22 |
Tachycardia | 1 | 3 |
Drug boluses | 5 | 15 |
Distraction | 4 | 12 |
Deformity correction | 5 | 15 |
Osteotomies | 3 | 9 |
Screw misplacement | 2 | 6 |
Deep anesthesia | 2 | 6 |
Electrodes disconnection | 1 | 3 |
Hypothermia | 1 | 3 |
Unknown | 2 | 6 |
Three patients (13.6%) had persistent alerts; sudden loss of MEPs in both lower limbs was seen in one patient following accidental injury to the spinal cord by pedicle sound through a misplaced screw tract and two patients had decreased MEPs from both lower limbs following distraction. Stagnara wake up test was performed in two patients, out of whom one patient had negative and one had a positive result, while the wake up test could not be performed in one patient. The surgeon decided to continue with surgery in the patient with negative wake up test and MEPs were taken at more frequent intervals; the MEPs restored to baseline value after 50 min and patient woke up with no new neurologic deficit. In the patient with positive wake up test, it was decided to abort the case at that stage and the patient woke up with postoperative deficit. Neurodeficit resolved after a duration of 4 months. Surgery was also aborted in third patient in whom the wake up test couldn’t be done, as the MEPs were persistently low even after all corrective measures had been instituted (Figure
No significant differences were noted in age and gender between patients with no alerts and those who had alerts with or without postoperative deficits. Electrodes were displaced during surgery in three patients, in one case electrodes were reinserted, while in other two cases, surgery was continued without reinsertion. We had no complications with TcMEPs during or after the surgery. All the alerts are shown in Table
S. no. ( |
Patient number ( |
Age (years) | Diagnosis | Type of alert | Intraoperative motor evoked potential recovery (Y/N) | Cause of alert | Wake up test performed (Y/N) | postoperative Neurological deficit (Y/N) | Recovery at final follow-up |
---|---|---|---|---|---|---|---|---|---|
1 | 1 | 2–4 | Congenital scoliosis | Parallel | Y | Hypotension | N | N | |
2 | Non-parallel | Y | Osteotomy | ||||||
3 | 2 | 10–12 | Idiopathic scoliosis | Non-parallel | Y | Tachycardia | N | N | |
4 | 3 | 12–14 | Kyphoscoliosis | Non-parallel | Screw misplacement | Complete | |||
5 | 4 | 4–6 | Congenital scoliosis | Non-parallel | Y | Drug bolus | N | N | |
6 | Non-parallel | Y | Deformity correction | ||||||
7 | 5 | 50–51 | Post traumatic kyphosis | Non-parallel | Y | Distraction | N | N | |
8 | 6 | 14–16 | Idiopathic scoliosis | Non-parallel | Y | Deformity correction | N | N | |
9 | 7 | 9–11 | Idiopathic scoliosis | Parallel | Y | Deep anesthesia | N | N | |
10 | Parallel | Y | Hypotension | ||||||
11 | 8 | 13–15 | Idiopathic scoliosis | Non-parallel | Y | Screw misplacement | N | N | |
12 | 9 | 6–8 | Congenital scoliosis | Parallel | Y | Drug bolus | N | N | |
13 | Non-parallel | Y | Hypotension | ||||||
14 | 10 | 3–5 | Congenital kyphosis | Non-parallel | Y | Osteotomy | N | N | |
15 | 11 | 15–17 | Idiopathic scoliosis | Parallel | Y | Hypotension | N | N | |
16 | 12 | 7–9 | Kyphoscoliosis | Parallel | Y | Hypotension | N | N | |
17 | Non-parallel | Y | Distraction | ||||||
18 | 13 | 14–16 | Idiopathic scoliosis | Non-parallel | Y | Deformity correction | N | N | |
19 | 14 | 14–16 | Idiopathic scoliosis | Non-parallel | Y | Deep anesthesia | N | N | |
20 | 15 | Idiopathic scoliosis | Non-parallel | Y | Electrodes misplacement | N | N | ||
21 | 16 | 9–11 | Idiopathic scoliosis | Non-parallel | Osteotomy | N | |||
22 | Non-parallel | Y | Drug bolus | ||||||
23 | Non-parallel | Y | Distraction | ||||||
24 | 17 | 4–6 | Congenital kyphosis | Parallel | Y | Hypothermia | N | N | |
25 | Non-parallel | Y | Deformity correction | ||||||
26 | 18 | 10–12 | Idiopathic scoliosis | Non-parallel | Distraction | N | |||
27 | 19 | 5–7 | Congenital scoliosis | Parallel | Y | Hypotension | N | N | |
28 | Parallel | Y | Drug bolus | ||||||
29 | 20 | 13–15 | Idiopathic scoliosis | Non-parallel | Y | Unknown | N | N | |
30 | Parallel | Y | Hypotension | ||||||
31 | 21 | 8–10 | Congenital scoliosis | Non-parallel | Y | Deformity correction | N | N | |
32 | Non-parallel | Y | Unknown | ||||||
33 | 22 | 6–8 | Congenital kyphosis | Parallel | Y | Drug bolus | N | N |
Sensitivity and Sp of TcMEP in deformity correction surgery were 100 and 96.6%, respectively, in our study. NPV and PPV were 100 and 33.3%, respectively (Table
New neurological deficit | No new neurological deficit | |
---|---|---|
TcMEP alert | 1 (true positive) | 2 (false positive) |
No TcMEP alert | 0 (false negative) | 58 (true negative) |
Sn | 100% | |
Sp | 96.6% | |
PPV | 33.3% | |
NPV | 100% |
There were no instances of tongue or lip lacerations, seizures, or any other complications during or after the surgery.
The purpose of IONM is to provide real-time assessment of spinal cord function during surgery that involves cord manipulation. Various mechanisms of spinal cord injury in deformity correction surgery are distraction, ischemia, and compression (
In TcMEPs, stimulus is delivered to the motor cortex from subcutaneously placed corkscrew electrodes, and potentials are recorded from electrodes placed in various bilateral muscle groups. The purpose of recording potentials from maximum number of possible muscles is to increase the Sn (
An 80% or greater decrease in the MEP amplitude to be taken as a criteria for “alert” was introduced by Langeloo et al. (
In all our cases, IONM with TcMEPs was done with strict adherence to anesthesia protocol (TIVA) and checklist. An alert not synchronous with any high-risk surgical maneuver could be likely due to various non-surgical factors and such an alert when not quickly identified and corrected could mislead and compel the surgeon to take unreasonable risk or to change the surgery plan. A checklist places emphasis on all the likely surgical and non-surgical factors that cause an alert and thus a checklist might not allow any potential risk factor to be missed and to mark an alert due to any cause as a FP alert. As the systemic state varies from time to time, baseline potentials obtained at the beginning of surgery may no longer be appropriate at later point. In the intraoperative period, MEP amplitude has high trial-by-trial variability (
Skinner et al. (
We had no FN alerts in our study, although a few case reports exist in literature (
IONM with TcMEPs alone is not without limitations; anesthetic and systemic changes produce high variability in amplitudes, inhalational agents decrease the effectiveness of stimulation, and muscle relaxants inhibit amplitudes from muscles and thus adherence to strict anesthesia protocol is important. Reliability of MEPs diminishes in patients with preoperative neurological deficits. IONM cannot detect abrupt loss of signals as in anterior spinal artery syndrome because this is an acute process. Although TcMEPs have been used without any complications by Schwartz et al. (
This study was not a prospective study, study population may not reflect all the deformities (in patients with NM scoliosis and history of epilepsy only SSEPs were used) and no comparison was done with multimodality monitoring. Following an alert, respective teams performed their roles almost simultaneously; hence, the exact cause of an alert may not have been identified all the times. But, as we have an ongoing protocol; with checklist, we think that the point mentioned in the records would most probably represent the cause of an alert.
In neurologically normal patients, IONM with TcMEPs is a safe and efficacious real-time monitoring system to warn of impending neurological injury at a reversible stage, thus providing a window of opportunity for intervention. Type of alert (parallel or non-parallel) can differentiate systemic and focal compromise. Following the checklist helps in systematically analyzing the potential cause of alert and appropriate action to be taken. Wake up test still remains a valuable monitoring tool in situations of persistent alerts and can help the surgeon in decision-making. Finally, prompt action and close coordination amongst surgeon, anesthetist, neurophysiologist, and operating room staff are required to reduce neurologic mishaps.
This is the standard operating practice; only a standard methodology is assessed in our study.
SA – Senior Consultant, Department of Spine Surgery; NP, PG – Spine Fellow, MK – Consultant, Department of Anesthesia.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.