Thermal Protection Strategies and Neuromonitoring during Ablation

Abstract

Advanced interventional pain management approaches seek to lesion neural targets to achieve desirable analgesia; however, equally important is preservation of motor and sensory function for regional bystander nerves. The topic of neuroprotection is also relevant for thermal ablation of metastatic bone tumors in the vicinity of neural structures. This report aims to provide an IR-directed framework of thermoprotective techniques available during thermal ablation.

Thermal ablation is one of the core modalities available to interventional radiologists seeking to provide advanced interventional pain management. CT guidance has enabled accurate probe placement to treat targets previously unreachable with traditional fluoroscopic pain management. As the complexity of treatment targets increases and the fragility of patients increases, the topic of neuroprotection is increasingly relevant in this domain.

The fundamental goal when protecting a nerve from injury is to keep the juxtaneural temperature between 10 and 45 °C ^{1 2 3 4 5 6 7 8 9 10 11} as detailed in the thermoprotective temperatures given in Table 1 . The term “juxtaneural” is specified because the usual form of temperature confirmation is with a thermocouple placed adjacent to the nerve. If temperatures in this space begin to reach the high or low limits, active thermal corrective measures will be needed.

Table 1. Relevant temperatures in the domain of thermal ablation and thermoprotection.

Goal temp range for thermoprotection	Effect	Biofeedback	Effect
10 to −20 °C	Neuropraxia; Sunderland Category 1 (minutes to weeks to recover)	10 °C	Mild motor loss
−20 to −100 °C	Reversible cryoneurolysis; Sunderland Category 2 (weeks to months to recover)	7 °C	Mild sensory loss
Below −100 °C	Irreversible cryoneurolysis; Sunderland Category 3–5	0–5 °C	Total sensory and motor loss

Tools of Thermoprotection

The three categories of action include passive thermoprotection, active thermoprotection, and mitigation.

Passive Thermoprotection

Passive thermoprotection refers to strategies that employ real-time physiologic, imaging, or temperature measurements to assess thermal risk to a critical structure. Among the tools:

• - Biofeedback is the simplest and perhaps the most reliable form of passive thermoprotection, but it relies on the patient's alert cooperation during the procedure. Sensory biofeedback refers to asking the patient to report when there is pain in a particular distribution that corresponds to the nerve that the operator is trying to protect. It is notable that with cryoablation, sensory biofeedback may be subtle and brief, compared with radiofrequency ablation (RFA) or microwave where biofeedback is more pronounced. Biofeedback relies on the nerve not receiving lidocaine (intentionally or due to unintentional spread) prior to the ablative steps. Biofeedback can also be utilized for motor function by asking the patient to perform the motor function in question. A reduction in ability to perform the motor function should prompt active neuroprotective steps.

• - Intraoperative neuromonitoring (IONM) refers to real-time neurophysiologic electroneurologic monitoring that is exclusively performed when patients are under total intravenous general anesthesia.

• - Thermocouples can be deployed to the zone of ablation to measure juxtaneural temperatures in real time. These are often very thin (22–27 gauges) and can fit coaxially through a 17-gauge needle which allows for simultaneous passive and active thermoprotection via delivery of fluid around the thermocouple (further detailed in the “Active Thermoprotection” section). Thermocouples are often embedded into ablation devices to provide real-time temperature monitoring, though it is important to discuss with the vendor whether the temperatures reported on the screen are actual temperatures or the temperatures of coolant fluids or gas leaving the ablation site. Finally, an ablation device with embedded thermocouples that is deployed to the region can be passively used as a thermocouple if it is not being used for ablation.

• - 0 °C line imaging refers to CT-guided cryoablation where the 0-degree line is visible as the ice ball expands. This can be difficult to visualize in normal bone, and easiest to visualize in fluid or lytic tumor densities. As this temperature falls below the 10 °C thermoprotective threshold, caution should be exercised when the ice ball approaches and active thermoprotective maneuvers should be considered.

• - Neurophysiologic intraoperative monitoring (NIOM, frequently referred to as neuromonitoring) is the real-time monitoring and mapping of critical neurologic structures during a surgical procedure. Historically, somatosensory evoked potentials (SEPs) were used to monitor sensory pathways during scoliosis surgeries to protect against injury to the spinal cord. Later, motor evoked potentials (MEPs) and electromyography (EMG) monitoring were introduced to monitor central motor pathways and peripheral nerves, respectively. All three—SEP, MEP, and EMG monitoring—can be used as a type of passive thermoprotection during interventional radiological procedures that pose a risk to the peripheral nervous system.

SEPs are obtained by electrically stimulating a sensory or mixed peripheral nerve and recording the ascending volley along various points of the nervous system. ¹² The impulse is transmitted through the dorsal column pathways, and the final waveform, representing activation of the primary sensory cortex, is recorded with electrodes placed on the scalp. Injury to the sensory pathways as they ascend the peripheral nerve (or thereafter) can be detected by a change in the SEP waveforms. SEP monitoring can be performed continuously during the surgery, as it does not cause significant movement and is considered “invisible” to the radiologist or surgeon. A change in the amplitude (usually 50% or greater) or prolongation in latency (typically greater than 10%) of the SEP waveforms is considered significant and indicative of neural injury (the so-called warning criteria). ¹³ Fig. 1 shows a normal SEP waveform and one that has had a significant change.

Fig. 1.

Normal and abnormal somatosensory evoked potential (SEP) waveforms. ( a ) Normal SEP waveform elicited after left tibial nerve stimulation. Each column shows a typical waveform; the arrow points to the waveform recorded from cortical sensory areas. The arrowhead points to the response recorded from the tibial nerve. Time is listed on the right-most column. Notice that there is no significant change in the waveforms as time progresses. ( b ) SEP waveform elicited after left tibial nerve stimulation. Notice that the waveform in the left-most column marked with an arrow (recorded from the cortical sensory area) decreases in amplitude over time, suggesting potential compromise to the relevant dorsal column pathway. Time is listed on the right-most column.

MEP monitoring assesses the corticospinal (motor) pathways. A high intensity train of electrical stimuli is applied through needle electrodes in the scalp over the motor cortex. ¹⁴ This results in activation of the corticospinal tract, which results in activation of anterior horn cells in the spinal cord. Thereafter, peripheral motor nerves are activated, leading eventually to the production of muscle MEP, which are recorded with needle electrodes in multiple muscle groups in the upper and lower limbs. MEP monitoring is typically performed when there is potential of injury to motor pathways in the spinal cord and brain. When an MEP is recorded, there is considerable movement of the patient, and this type of monitoring is not “invisible” to the radiologist or surgeon. Thus, it can only be performed intermittently. A reduction of amplitude of greater than 90% in the case of spine surgery or greater than 50% in the case of brain surgery is considered the warning criteria and indicative of possible injury to the motor pathways. The MEP warning criteria in the case of peripheral nerve surgery is less well defined, but a 50% amplitude reduction is often used. ¹⁵ Fig. 2 shows a normal MEP and one that has had a significant change.

Fig. 2.

Normal and abnormal motor evoked potential (MEP) waveforms. ( a ) MEP obtained after stimulation of the scalp and recorded from several muscle groups in the left upper and lower extremities. Each column shows an MEP response from a single muscle group. Time is listed on the right-most column. Notice that there are no significant changes in the waveforms in any of the columns as time progresses. (Names of muscle groups noted in upper left hand of the columns.) ( b ) MEP obtained after stimulation of the scalp and recorded from several muscle groups in the right upper and lower extremities. Each column shows an MEP response from a single muscle group. Time is listed on the right-most column. Notice a gradual decrease in the MEP amplitude from the right foot, marked with an arrow. The loss of MEP amplitude suggests possible compromise of the motor fibers supplying the relevant muscle group. (Names of muscle groups noted in upper left hand of the columns.)

EMG monitoring is performed with needle electrodes inserted into muscles of interest. These muscles are selected based on the neural structures that are at risk during the surgery. ¹⁶ In the case of surgery on peripheral nerves, muscles innervated by the relevant nerve should be monitored. Additionally, a few nearby muscles should also be monitored to serve as controls. EMG monitoring typically consists of recording ongoing “spontaneous” activity. Injury to a peripheral nerve causes injury potentials, called neurotonic discharges, which can be detected by the needle electrodes in the target muscle. When these discharges are seen, they serve as a warning criterion for impending injury to the peripheral nerve. Additionally, in some cases, the peripheral nerve can be stimulated by an electrical stimulus and a compound muscle action potential (CMAP) recorded from the target muscle (often referred to as triggered EMG). Such stimulation can be used to identify neural structures in the operative field and it can also be used to determine the distance of the nerve from the stimulating electrode, based on the stimulation intensity needed to produce the activation. This type of monitoring is frequently used in peripheral nerve surgery to minimize injury to the nerve and can be performed continuously.

When SEP and MEP monitoring are used, anesthetic protocol may need to be altered to optimize these responses. Often total intravenous anesthesia with propofol and an analgesic (i.e., remifentanil) are used. ¹⁷ However, if only EMG monitoring is needed, it is relatively insensitive to anesthetics, and the use of inhalational anesthetics is compatible with successful EMG monitoring. However, paralytics cannot be used when EMG monitoring is needed as that will prevent the recording of neurotonic discharges and stimulated CMAPs.

Neuromonitoring can serve as a useful tool in passive thermoprotection during IR procedures. The protocol used will be dictated by the structure undergoing thermal ablation and the neural tissue that is at risk. If it is a peripheral nerve, neuromonitoring may involve recording SEP after stimulating that nerve and recording MEP from muscles innervated by it, if possible. EMG can also be monitored from the same muscles being used to record MEP. If a plexus is at risk, monitoring multiple nerves with SEP and multiple muscle groups with MEP and EMG may be necessary. In some cases, depending on the nerve or neural structure at risk, only EMG monitoring may be possible.

A team of professionals is usually needed for successful neuromonitoring. A neuromonitorist is responsible to the technical aspects of the monitoring, while a clinical neurophysiologist interprets the data and supervises the neuromonitorist. In all cases, a discussion between the radiology, anesthesia, and neuromonitoring team before the procedure is important to determine the most appropriate monitoring and anesthetic regimen. Constant and open interaction between the teams during the procedure is critical to ensure successful monitoring and optimal outcomes.

Active Thermoprotection

Active thermoprotection refers to maneuvers performed to change the temperature in a region of interest. Among these maneuvers:

• - Pneumodissection refers to instilling carbon dioxide or room air ¹⁸ which have an order of magnitude less thermal conductivity (0.014–0.024 for gas vs. 0.56 W/mK for water) to insulate a protective structure. Compared with fluid, gas will seek nondependent anatomy. Carbon dioxide will tend to be absorbed within minutes and therefore touch-up injections will be periodically necessary throughout the procedure. In contrast, room air is more durable but if injected intraperitoneally may be more symptomatic than carbon dioxide. Carbon dioxide is filtered, while room air may be double HEPA filtered in most operating theaters. Due to their high effectiveness at low volumes, pneumodissection techniques have become a mainstay in thermoprotection of the spinal cord during vertebral RFA and cryoablation procedures, where often a volume of 10 to 40 mL ¹ in total is sufficient. Based on adhesions or fascial planes, the gas spread may be unpredictable in the craniocaudal axis and additional scanning may be needed to fully confirm the area of spread of the gas. Finally, because gas is not sonolucent, this method is most useful during CT-guided procedures and specifically less useful in ultrasound-guided procedures.

• - Hydrodissection refers to the targeted instillation of fluid to an area where ablation will be performed in an effort to provide a thermal barrier to the protective structure. Often for CT-guided procedures, the fluid of choice will be normal saline mixed with iodinated contrast, but for RFA procedures (especially monopolar) the fluid of choice will be dextrose to avoid having a plasma field develop and nontarget electrical conduction. Hydrodissection is typically performed via a Yueh, Gangi, or trocar needle placed into a region of interest prior to the ablation. Spread of fluid may be affected by adhesions from prior radiotherapy or surgery.

• - Hydroconvection refers to a consistent flow of protective fluid injected into a space that is too small to accommodate a sufficient volume of hydrodissection fluid, or an area that cannot expand due to extreme scar tissue. One such time is during ablation of targets in the neck which are relatively superficial, during which a needle can be placed subcutaneously to protect the skin with hydroconvection. The ice ball is visualized well under ultrasound as the continuously renewed fluid buffer provides an avid sonographic window.

Mitigation

Finally, mitigation refers to the strategies to reduce injury to the nerve prophylactically or once injury has occurred. Thermal injury to the nerve is characteristically a neuritis, with varying degrees of injury based on the time and duration of thermal exposure. In general, cryoablation provides a greater chance for eventual nerve recovery compared with RFA and microwave where frank denaturation of the nerve occurs. The mainstay of mitigation is intravenous (IV) steroids. Dexamethasone (e.g., up to 10 mg of dexamethasone preservative free) can be injected directly onto the nerve of interest. IV steroids, such as dexamethasone (Decadron) 8 to 10 mg IV, can be given before a high-risk procedure or at the first sign of nerve injury. Treatment can be continued in the outpatient setting with a steroid taper. Nerve recovery may be expected to begin at around 6 months after injury.

Conclusion

Thermoprotection remains an important topic as interventional radiologists perform increasingly complex interventional pain management procedures on increasingly fragile patients. A multitude to passive and active thermoprotective techniques are available but ultimately the most important strategy involves a knowledge of the nerves in the region, and a thoughtful informed consent which prepares the patient for potential unwanted nerve issues postprocedure. A more favorable nerve recovery profile of cryoablation may lean operators more toward that modality when possible.

Funding Statement

Funding No funding was received for any portion of this work.