Percutaneous Computed Tomography-Guided … Computed Tomography-Guided Spinal Destructive Procedures for Pain ... he spinal cord, ... surgery had to be performed under general anesthesia,

ORIGINAL ARTICLE

Percutaneous Computed Tomography-GuidedSpinal Destructive Procedures for Pain Control

Yucel Kanpolat, MD,* Ali Savas, MD, PhD,* Serdar Akyar, MD,† and Eric Cosman, PhD‡

Abstract: The techniques and results of the main percutaneous pro-

cedures performed using image guidance with computerized tomog-

raphy, including cordotomy, trigeminal tractotomy-nucleotomy, and

extralemniscal myelotomy, are presented. The advantages of image

guidance for anatomical orientation during the procedures and the

newly designed radiofrequency needle-electrode systems are stressed.

Key Words: computed tomography guidance, computed tomogra-

phy electrodes, image-guided surgery, pain, cordotomy, trigeminal

tractotomy, extralemniscal myelotomy

(Neurosurg Q 2004;14:229–238)

The spinal cord, which carries the pain-transmitting nervefibers in tracts, is a key component of the central nervous

system and an important area for pain surgery. A critical stepfor pain surgery was the first open cordotomy in a humanpatient, carried out by Martin at Spiller’s instigation in 1911.1

For many years thereafter, surgical interruption of the pain-transmitting tracts was performed using open techniques forintractable chronic pain management. These procedures werethe ‘‘gold standards’’ of their time but had several disadvan-tages in practice, including the following: 1) open surgery couldnot be tolerated by patients in the terminal stage or with a poorinternal status; 2) surgery had to be performed under generalanesthesia, which precludes cooperation with the patient andmonitoring of neurophysiologic functions; and 3) irreversiblelesions were made without any control of function, making theprocedures rather risky.

The high mortality and morbidity risks of opencordotomy prompted neurosurgeons to develop a less invasiveprocedure using a percutaneous approach. The first applicationof a percutaneous procedure on the spinal cord with a radio-active strontium needle was described by Mullan et al2 in 1963and is considered the most important contribution in theadvancement of percutaneous (minimally invasive) techniquesin pain surgery. Percutaneous techniques solve the prob-lems described previously but require new technology suchas imaging techniques, radiofrequency (RF) systems for

neurophysiologic testing and lesioning, and needle-electrodesystems.

COMPUTED TOMOGRAPHY GUIDANCE FORSPINAL CORD SURGERY

Percutaneous procedures require an imaging system tovisualize the target-electrode relation. Initially, x-ray andpositive-contrast myelography provided indirect informationvia visualization of the dentate ligaments as well as the anteriorand posterior parts of the spinal cord. Indeed, the most impor-tant disadvantage of conventional percutaneous stereotacticpain procedures was in the imaging technique used, because itwas not possible to demonstrate directly the real shape anddiameters of the spinal cord or the target-electrode relationsdespite the use of contrast agents.3,4 Although it is an effectiveprocedure, the technique’s shortcomings resulted in highermorbidity than expected in the reported series and, conse-quently a decrease in its popularity in intractable pain surgeryin the past 20 years. This problem could be solved only byusing computerized tomography (CT) as a direct imagingtechnique instead of x-ray myelography. In the past decade,CT-guided percutaneous spinal cord procedures have beenpresented as effective and safe methods of destroying thepain-transmitting tracts for the treatment of patients with intrac-table chronic pain.

Demonstration of the needle-electrode system with CTscans provides information about the location of the electrodetip in the spinal cord. Impedance measurements give directinformation about the compartment in which the tip is locatedin the cerebrospinal fluid (CSF) or the spinal cord.5,6 Electricalstimulation of the target is the most important aspect of thisfunctional neurosurgical operation. A functional and anatom-ical description of the target is possible only after taking thesteps described previously.

CT guidance offers the advantage of topographic orien-tation to the spinal cord, allowing selective denervation. Withthe help of this visualization system, spinal cord diameters canbe measured for each patient individually, the target-electroderelation can be assessed directly, and the needle-electrode sys-tem can be inserted into the specific part of the pain-conductingsystem necessary for achievement of selective cordotomyand selective trigeminal tractotomy-nucleotomy (TR-NC).7–10

Ablative spinal cord procedures have thus become trulystereotactic operations as well as being functional neurosur-gical operations. The described techniques must be performedwith the newly designed RF needle-electrode system, however,because these electrodes make it possible to perform selective

From the *Department of Neurosurgery, Ankara University School of Medicine,Ankara, Turkey; †Department of Radiology, Ankara University School ofMedicine, Ankara, Turkey; and ‡Department of Physics, MassachusettsInstitute of Technology, Cambridge, MA.

Reprints: Yucel Kanpolat, MD, Inkilap Sk. 24/2, Kizilay 06650, Ankara,Turkey (e-mail: [email protected]).

Copyright � 2004 by Lippincott Williams & Wilkins

Neurosurg Q � Volume 14, Number 4, December 2004 229

JOBNAME: nq 14#4 2004 PAGE: 1 OUTPUT: Mon November 1 18:49:12 2004

lww/nq/88532/04-123

cordotomy or selective trigeminal TR-NC with direct visuali-zation of the spinal cord.

RADIOFREQUENCY SYSTEMS ANDCT-COMPATIBLE ELECTRODES

Many techniques have been used to destroy nervoustissue selectively. After the work of Sweet and Mark,11 RFcurrent began to gain wider acceptance because of its advan-tages in creating more controllable neural lesions. RF systemshave become a standard tool in selective lesioning of thenervous system because they facilitate controlled lesioningand provide valuable information about the impedance valuesand electrical stimulation of the tissue in which the electrode islocated.12,13 In addition, with patient cooperation in responseto neurostimulation, the functional features of the spinal cordregion in which the needle-electrode system is placed areeasily evaluated for correct positioning of the lesion.14,15

In 1965, Rosomoff et al15 described the technique ofpercutaneous cordotomy with an RF needle-electrode system.Since then, techniques involving RF thermic lesions have beenthe standard way to perform percutaneous pain procedureswithin the spinal cord.2,15,16–20 A subsequent development wasthe invention of thermocouple electrodes, which facilitatecontrol of the temperature of the tissue surrounding the tip.Standard RF needle-electrode systems are used with x-rayvisualization, however, and are not compatible with CT scans.

The authors have designed an RF needle-electrodesystem with an RF generator (KCTE electrode kit and RFG-3C generator; Radionics, Burlington, MA). For this system,20-gauge thin-walled needles with plastic hubs were designedto avoid imaging artifact problems.21 These needles havedemarcations on the shaft beginning 4 cm from their tips toshow the depth of insertion to support metrically the sensationof penetration (ie, the distal part of the needle is marked every1 cm up to 4 cm, and the proximal part is marked at 0.5 cm).Included in the kit are 4 different 2-mm open-tipped ther-mocouple electrodes: 0.30 and 0.40 mm in diameter with astraight tip and 0.30 and 0.40 mm in diameter with a curvedtip. The tip of the curved electrode has a 0.5-mm axis, whichallows better direction and manipulation of the electrode tip toreach the ideal target. The proximal part of the electrode haslines every millimeter for 6 mm, and a sizing clamp is attachedto this part (Fig. 1). The depth of the inserted part (3–4 mm) ofthe straight and curved active electrodes must be adjustedbefore application according to the measured diameters of thespinal cord. The screw of the sizing clamps is tightened in thesame direction as the tip of the curved electrode to facilitateawareness of the direction of the curved tip.

CT-GUIDED PERCUTANEOUS PROCEDURESCT-guided percutaneous pain procedures on the spinal

cord comprise the following: cordotomy, trigeminal TR-NC,and extralemniscal myelotomy (ELM).

CT-Guided Percutaneous CordotomySpinothalamic cordotomy aims to interrupt the spino-

thalamic tract (STT) ascending contralaterally to the painful

side, because most of the pain fibers decussate over 2 to 5 seg-ments before entering the STT.22,23 The anterolateral afferentsystem has a somatotropic relation with fibers from higherlevels laminating medially and ventrally.22,24 The topographicrepresentation within the pain tracts usually places the sacralsegments more posterolaterally within the segments, whereasthe cervical segments are situated more medially and ante-riorly. This segmentation allows selective cordotomy to achieveselective denervation of the upper or lower part of the body.

The most important structures surrounding the STT thatshould be considered during the procedure may be summa-rized as follows: 1) the descending autonomic pathwaysintermingling with the reticulospinal tract for vasomotor andgenitourinary control as well as mediation of respiration,located anteromedially; 2) the corticospinal tract as the majorpathway for motor control of the muscles, located posteri-orly; and 3) the ventral spinocerebellar tract for coordinationof movement of the extremities, which overlies the lateralSTT.22,25 Complications after cordotomy are related to theseanatomical relations. Bilateral lesions within the anterior partof the STT may cause sleep-induced apnea as a seriouscomplication resulting from bilateral lesioning of thereticulospinal tract.26–32 Therefore, bilateral selective dener-vation of the upper part of the body via cordotomy is contra-indicated. The sacral segment (posterolateral) of the STT isnot far from the pyramidal tract; thus, ipsilateral hemi- ormonoparesis may occur, being mostly transient in nature, withselective cordotomy.24,33,34 Cardiovascular or urinary distur-bances may also be expected. Cordotomy is indicated espe-cially for patients suffering from unilateral localized cancerpain (eg, pulmonary malignancies and mesothelioma), but it isalso appropriate for a wide range of patients, because theprocedure is much safer and more selective than conventionalcordotomy. CT-guided cordotomy may be recommended topatients even before the use of narcotics, because the proce-dure reduces patients’ dependence on the hospital, allowingthem to return to their previous daily activities. The proceduremay be performed bilaterally in patients suffering from bi-lateral lower body or low back pain caused by malignancy,whereas bilateral conventional cordotomy may be ratherrisky.9,35 Additionally, nonmalignant intractable somatic painmay be relieved by CT-guided percutaneous cordotomy, but allother treatment methods should be performed before surgery.

CT-guided percutaneous cordotomy was performed in197 cases at our clinic. One hundred eighty-five patients weresuffering from intractable unilateral pain caused by malig-nancy. Pulmonary malignancies (56 cases), mesothelioma (23cases), and Pancoast tumor (14 cases) represented most of thecases in our series. Other patients included 20 with gastro-intestinal carcinoma, 22 with metastatic carcinoma, and 50with multiple malignancies. The procedure was also applied to13 patients with benign pain, whereas 13 patients with bilateralcancer pain of the lower trunk and lower extremities weretreated with CT-guided bilateral selective cordotomy.

Surgical Technique

Patients should fast for 5 hours before surgery. The con-trast agent (7 mL Iohexol at a rate of 240 mg/L) is administered

230 q 2004 Lippincott Williams & Wilkins

Kanpolat et al Neurosurg Q � Volume 14, Number 4, December 2004


lww/nq/88532/04-123

via lumbar puncture 20 minutes before the operation. Thepatient should be kept in the Trendelenburg position beforeadmission to the CT unit. Alternatively, if a lumbar punctureis not possible, the contrast agent (5 mL Iohexol at a rate of240 mg/L) can be introduced laterally at the C1 to C2 level atthe beginning of the procedure. Patients should be awake andcooperative throughout the procedure to facilitate observationof neurologic functions, and no general anesthesia shouldbe given.

CT-guided percutaneous cordotomy is performed in theCT unit with the patient in the supine position. The head isflexed and fixed. Plastic-hubbed 20-gauge needles speciallydesigned for CT-guided procedures are used. Thinner needlesare not recommended, because directing the needle may bemore difficult with manipulation. Before needle insertion, alocal anesthetic agent is given and the neck is punctured withthe cordotomy needle just below the tip of the mastoid processin a vertical plane. Placement of the needle at the C1 to C2level can be seen in the lateral scanogram (Figs. 2, 3), and the

direction of the needle can be manipulated toward the anteriorpart of the spinal cord with the help of axial CT sections toobtain topographic localization. Axial CT scans with a 1-mmslice thickness are obtained. Target-needle and -electrode rela-tions can easily be visualized using CT guidance. Diametralmeasurements of the spinal cord are taken, and the activeelectrode is adjusted accordingly. The needle is inserted in thelateral axis, and the direction is corrected in accordance withCT scans (Figs. 4, 5). The needle is in the ideal position if it isnearly perpendicular to the spinal cord. The target in percu-taneous cordotomy is the lateral STT in the anterolateral partof the spinal cord at the C1 to C2 level. The electrode is in-serted into the target after achieving the ideal position of theneedle tip toward the target. If the cord is not punctured, thiscan be determined by means of CT scans (Fig. 6) and imped-ance measurement, and the electrode can be reinserted. Finally,the electrode tip is placed in the STT (Figs. 7, 8).

Impedance measurements are important for identifyingwhether the active electrode tip is in contact with the spinal

FIGURE 1. Kanpolat electrode kit(Radionics, Burlington, MA) cannulaand electrodes.

q 2004 Lippincott Williams & Wilkins 231

Neurosurg Q � Volume 14, Number 4, December 2004 Spinal Destructive Procedures for Pain Control


lww/nq/88532/04-123

cord or in the spinal cord. The measurements are an importantindication of passage into a new medium along the path of theneedle electrode. Impedance values are approximately 400 Vin the CSF; an increase of approximately 200 V is observedwhen there is contact between the electrode tip and the pia.The impedance value is almost always greater than 1,000 Vafter insertion into the cord.

Stimulation provides direct neurophysiologic informa-tion about the location of the electrode tip. Electrical stimu-lation with low (2–5 Hz, 0.3–1 V) and high (50–100 Hz, 0.2–0.3 V) frequencies are used. Stimulation with low frequenciescauses ipsilateral trapezius muscle contractions, indicating thatthe electrodes are within or near the anterior gray matter.In high-frequency range stimulation (50–60 Hz, 0.5–3 V),patients describe sensations such as contralateral tingling.

The final step is lesioning of the target area. During le-sioning, the energy and tip temperature are increased graduallyand ipsilateral motor functions must be tested continuously.We use 0.4-mm diameter electrodes for unilateral cordotomy

and 0.3-mm diameter electrodes for bilateral cordotomy. Thefinal lesion is made at 70�C to 80�C for 60 seconds. Usually, 2(maximum of 3) lesions are made in unilateral cordotomy. Ifrequired, bilateral staged procedures can be performed witha 1-week interval.

The patient should be hospitalized for 1 night after theprocedure. Neurologic examination is repeated after surgery,and when a bilateral cordotomy is performed, blood pressureand respiratory functions must be carefully observed andmonitored.

Results

The initial success rate of CT-guided percutaneouscordotomy was 95.5% (188 of 197 cases). In 153 cases(77.6%), only the painful region of the body was relieved ofpain, thus achieving selective cordotomy. Nonselective paincontrol was achieved in 32 cases (16.6%). Total pain control

FIGURE 3. Needle shown at the C1 to C2 level (arrow) ona lateral scanogram.

FIGURE 4. Displacement of the needle too anteriorly on anaxial computed tomography scan.

FIGURE 2. Representation of theneedle at the C1 to C2 level duringa cordotomy procedure.




lww/nq/88532/04-123

was achieved in all 13 cases (100%) treated with bilateralselective cordotomy.

No mortality or persistent morbidity was encountered inour series. Five cases of motor dysfunction, 3 cases of ataxia,1 case of urinary retention, and 1 case of hypotension wereobserved as temporary complications. Three cases of post-cordotomy dysesthesia and 1 case of Horner syndrome werealso observed.

CT-Guided Trigeminal Tractotomy-NucleotomyManagement of certain craniofacial pain conditions, such

as anesthesia dolorosa, atypical facial pain, pain caused bycraniofacial malignancies, intractable trigeminal neuralgia,geniculate neuralgia, glossopharyngeal neuralgia, and post-herpetic neuralgia, presents a great challenge to physicians.CT-guided percutaneous TR-NC is an important image-guidedpercutaneous destructive procedure on the spinal cord for thetreatment of such conditions.10,36

The nociceptive fibers from the fifth, seventh, ninth, andtenth cranial nerves descend in the medulla, forming the de-

scending trigeminal tract. These fibers ramify in the trigeminalnucleus caudalis near their point of termination.18,37–41 Thespinal trigeminal nucleus extends caudally through the me-dulla into the spinal cord to the C2 level.42,43 The topographiccollective organization of nociceptive fibers from the cranialnerves has made this region the most eloquent target forneurosurgeons.16,18,36,37,44–47 Operations such as trigeminalTR, NC, and nucleus caudalis dorsal root entry zone (DREZ)lesioning have been described and performed with variousopen and percutaneous techniques to interrupt the nociceptivefibers from the fifth, seventh, ninth, and tenth cranial nerves.The described surgical techniques still have disadvantages,however, with their high complication rates and unstablepostoperative results.

The indications for CT-guided percutaneous trigeminalTR-NC are similar to those for the nucleus caudalis DREZoperation, but the effectiveness of the procedure is high andthe complication rate is extremely low.10,36,42,48 In addition,trigeminal TR-NC is safe and effective for repeated appli-cations. Moreover, the DREZ operation may be performed ifTR-NC is ineffective or partially effective. With its advantages

FIGURE 5. Correct and final position of the needle on an axialcomputed tomography scan.

FIGURE 6. The unpunctured and displaced spinal cord on anaxial computed tomography scan.

FIGURE 8. Final position of the electrode (arrow) in theanterolateral spinal cord.

FIGURE 7. Schematic drawing of the needle-electrode systemplaced in the anterolateral spinal cord and pertinent anatom-ical structures.




lww/nq/88532/04-123

of minimal invasiveness, high success rate, and low compli-cation rate, we propose that percutaneous CT-guided TR-NCbe considered as a rational alternative to other pain-relievingprocedures.

The present series included 53 patients suffering fromatypical facial pain (16 cases), pain caused by craniofacial ma-lignancies (11 cases), glossopharyngeal neuralgia (13 cases),multioperated trigeminal neuralgia (3 cases), geniculate neu-ralgia (4 cases), postherpetic neuralgia (4 cases), and anesthe-sia dolorosa (1 case).

Surgical Technique

A temperature-monitoring RF electrode system designedspecifically for CT-guided TR-NC (KCTE electrode kit) wasused. Iohexol (7 mL) was administered into the subarachnoidspace by lumbar puncture 20 minutes before the procedure.

The patient was placed in the prone position on the CT table,and the head was flexed and fixed. After a lateral scanogram(Fig. 9), axial CT scans were obtained in the C1-occiput regionand the cord diameters were measured. The active electrode tipwas adjusted according to the diametral measurements.

A 20-gauge needle was inserted from the C1-occiputregion 5 to 7 mm lateral to the midline (Fig. 10). The targetwas 3 mm anterior to the posterior aspect of the spinal cord and5 to 6 mm lateral to the midline at the first cervical segment(see Fig. 9). Correct placement of the needle on the target wasfacilitated by CT visualization. After final positioning of theneedle (Fig. 11), a 2-mm, open-tipped, straight or curvedelectrode was inserted into the needle. The final position of theelectrode tip was confirmed with new CT scans and impedancemeasurements (Figs. 12, 13). As the most important param-eter, electrical stimulation was used to confirm the localizationof the electrode. Paresthesia of the ipsilateral half of the facecan be observed with stimulation in most cases, indicating thatthe tip of the electrode is in the nucleus tract complex. Ifparesthesia of the face occurs, slight withdrawal of the tip andrestimulation may cause a dysesthetic sensation in the throat orinside the ear, indicating that the tip is in the nociceptive fibersof the seventh, ninth, and tenth cranial nerves. RF lesions (1–3in number) were made at 65�C to 75�C for 60 seconds.

Results

The best results were obtained from patients with glosso-pharyngeal neuralgia (13 cases) and geniculate neuralgia (4cases). Pain was effectively controlled in 11 of these cases.

TR-NC showed promising results in patients sufferingfrom pain caused by craniofacial malignancies, and pain con-trol was achieved in 9 (82%) of the 11 patients.

In 7 of 16 patients with atypical facial neuralgia, painwas controlled effectively. A nucleus caudalis DREZ operation

FIGURE 9. Needle placement on a lateral scanogram.

FIGURE 10. Insertion of the needle intothe C1-occiput level (arrow) in the proneposition.




lww/nq/88532/04-123

was required for 2 patients after TR, however, and pain controlwas obtained in only 1 of them after a DREZ operation. Painwas controlled in 2 patients with postherpetic neuralgia. In the3 patients suffering from anesthesia dolorosa, TR-NC wasineffective. Pain control was obtained in 2 of 3 intractabletrigeminal neuralgia patients, whereas an additional DREZoperation was partially effective in 6 patients. Postoperativetemporary ataxia was the only complication observed in 5cases, and no mortality was observed.

CT-Guided Extralemniscal MyelotomyELM is a procedure performed at the cervicomedullary

junction of the spinal cord and consists of central cord le-sioning for intractable pain.17,19,49,50 The neurophysiologicmechanism of pain relief after ELM remains unclear. Theprocedure was developed after empiric observations made byHitchcock that loss of pain in the lower body is the result ofan entirely separate pure pain pathway ascending the spinalcord that is possibly nonspecific, multisynaptic, and close to

the midline. Gildenberg and Hirshberg51 performed a limitedmyelotomy with an open technique at the T10 level to obtainpain control in the lower part of the body.

Ideal patients for the procedure are those with intractablepain located in the midline or with bilateral involvement. Aseries of 16 patients (15 with intractable cancer pain and 1 withpain caused by traumatic paraplegia) were managed by CT-guided, percutaneous ELM in our department. In all cases,pain was located in the lower trunk bilaterally and/or at themidline.

Surgical Technique

The operative technique is rather similar to CT-guidedcordotomy and TR-NC. All patients were awake and cooper-ative throughout the procedure to facilitate observation ofneurologic functions. Contrast material (7 mL Iohexol) wasadministered into the subarachnoid space via lumbar puncture20 minutes before the operation, after which the patient waskept in the Trendelenburg position for 5 minutes. The patientwas then placed on the CT table in the prone position.

Two types of electrodes of different sizes were used:Rosomoff cordotomy electrodes with a diameter of 0.45 mmand a 1-mm open tip (RCK-2A; Radionics) and, for 8 patients,a special needle-electrode system for CT-guided procedures(KCTE electrode kit). Diametral measurements of the spinalcord were taken using the CT scans at a 1-mm slice thickness,and the active electrode was adjusted accordingly. The needlewas inserted at the C1-occiput level through the midline (Fig.14). Placement of the needle at the C1-occiput level can beseen in the lateral scanogram (Fig. 15), and the direction ofthe needle can be manipulated toward the posterior part of thespinal cord with the help of axial CT sections (Fig. 16). Thetarget in ELM is the center of the spinal cord at the C1-occiputlevel. When the needle was approximated just behind the cord,the adjusted active electrode tip was inserted into the cord viathe needle (Figs. 17, 18). Impedance measurements weretaken, and stimulation was given for physiologic control of theposition of the active electrode. Paresthesia of the distal lower

FIGURE 11. Final position of the needle on an axial computedtomography scan.

FIGURE 12. Schematic drawing of the needle-electrode systemplaced in the posterolateral spinal cord for a tractotomy-nucleotomy and pertinent anatomical structures.

FIGURE 13. Final position of the electrode in the posterolateralspinal cord for a tractotomy-nucleotomy on an axial computedtomography scan.




lww/nq/88532/04-123

limb was observed with stimulation (50–75 Hz, 0.2–0.5 V) inmost of our cases. A controlled lesion was made at 70�C to80�C for 60 seconds, and 2 lesions were usually made.

Results

In six of the cases with malignancy (42.8%), total painrelief was achieved, partial satisfactory pain relief wasobtained in 4 cases (28.5%), and no pain control was achievedin 4 cases (28.5%). The procedure was partially effective in theparaplegic patient. No complications were observed.

In the group of 8 patients in whom the larger electrodes(0.45-mm diameter with 1-mm open tip and 0.25-mm diameterwith 2-mm open tip) were used, total pain relief was achievedin 5 patients (62.5%) and partial satisfactory pain relief wasachieved in 2 patients (25%), although pain persisted in 1 patient(12.5%). Conversely, in the group of 6 patients in whom the

smaller electrodes (0.25-mm diameter with 1-mm open tip)were used, total pain relief was achieved in 1 patient (16.6%)and partial satisfactory pain relief was obtained in 2 patients(33.3%), although pain persisted in 3 patients (50%). Thus, theresults of procedures performed with the smaller electrodeswere poor, causing us to change our electrode dimensions(0.4-mm diameter and 2-mm open tip).

CONCLUSIONThe spinal cord is an ideal target for percutaneous stereo-

tactic pain procedures. Extensive use of morphine pumps andelectrode stimulating systems has not been the main reason forthe decline in popularity among neurosurgeons of percutane-ous cordotomy and destructive pain procedures on the spinalcord in intractable pain surgery over the last 2 decades. The

FIGURE 15. Needle placement on the lateral scanogram.FIGURE 16. Final position of the needle on an axial computedtomography scan.

FIGURE 14. Insertion of the needle into theC1-occiput level for an extralemniscalmyelotomy in the prone position.




lww/nq/88532/04-123

high mortality and morbidity of the percutaneous proceduresperformed in the past were the most important reasons for thisneglect. It was not possible to describe the conventional per-cutaneous techniques as stereotactic, because the term means3-dimensional localization in space. Destructive percutane-ous procedures on the spinal cord have conventionally beenperformed with the aid of positive x-ray visualization, and itwas not possible to determine the diameters, target area, orlocalization of the cord. These disadvantages caused highercomplication rates and poor outcomes and thus impeded thecommon use of this surgical technique.

CT guidance provides a new perspective in destructivepain surgery on the spinal cord because of its dynamic pro-perties and capabilities. CT made the procedures truly stereo-tactic and image guided. The spinal cord diameters can be

measured for each patient individually, and the cord can bevisualized directly.8,35,36,49 The most important advantage ofCT guidance is that it allows visualization of the activeelectrode location in the spinal cord. Additionally, the tip of theactive electrode may be redirected if needed during theprocedure, based on the dynamic images. Under CT guidance,displacement and rotation of the spinal cord are visualized andthe surgical strategy of the procedure can be correctlyreoriented. These advantages are the result of contemporaryadvances in image-guided stereotactic surgery. These areeffective and relatively inexpensive operations. Patients are notdependent on physicians or devices; consequently, they mayreturn to their active lives after complete pain control isachieved.

Pain remains a significant problem in this century despiteprofound medical, technologic, and pharmacologic advances.There is a large population of patients who may be good candi-dates for this inexpensive, noninvasive, and highly effectiveprocedure. Worldwide, 6 million people die annually asa result of malignancies. In the terminal stage of malignancy,60% to 90% of patients experience intractable pain and 20% ofthese patients remain untreated despite modern technology. Asan example, in the United States alone, 1,382,400 people arediagnosed with cancer annually, whereas 160,400 die ofpulmonary carcinoma and 560,000 as a result of malignan-cies.52–54 Unilateral cordotomy may help these patients in theirterminal stage. The indications for CT-guided TR-NC inintractable pain in the craniofacial region are almost the sameas for the nucleus caudalis DREZ operation, but the effec-tiveness of the procedure is remarkably high and the compli-cation rate is extremely low. CT-guided percutaneous ELM isa neglected but valuable procedure in the treatment of visceralpain or cancer pain located in the midline and/or bilaterally.

In conclusion, CT-guided pain procedures should beused efficiently by neurosurgeons in the treatment of intrac-table pain, a practice they seem to have forfeited to other disci-plines during the past 2 decades.

ACKNOWLEDGMENTSThe authors express gratitude to Ahmet Sinav, MD, for

creative drawings and to Helen Stevens for editing themanuscript.

REFERENCES1. Spiller WG, Martin E. The treatment of persistent pain of organic origin in

the lower part of the body by division of the antero-lateral column of thespinal cord. JAMA. 1912;58:1489–1490.

2. Mullan S, Harper PV, Hekmatpanach J, et al. Percutaneous interruption ofspinal pain tracts by means of a strontium 90 needle. J Neurosurg. 1963;20:931–939.

3. Gildenberg PL, Lin PM, Polakoff PP, et al. Anterior percutaneous cervicalcordotomy. Determination of target point and calculation of angle ofinsertion. Technical note. J Neurosurg. 1968;28:173–177.

4. Onofrio BM. Cervical spinal cord and dentate delineation in percutaneousradiofrequency cordotomy at the level of the first to second cervicalvertebrae. Surg Gynecol Obstet. 1971;133:30–34.

5. Gildenberg PL, Zanes C, Flitter M, et al. Impedance measuring device fordetection of penetration of the spinal cord in anterior percutaneouscervical cordotomy. Technical note. J Neurosurg. 1969;30:87–92.

6. Taren JA, Davis R, Crosby EC. Target physiologic corroboration instereotactic cervical cordotomy. J Neurosurg. 1969;30:569–619.

FIGURE 18. Final position of the electrode in the posterolateralspinal cord for an extralemniscal-myelotomy on an axialcomputed tomography scan.

FIGURE 17. Schematic drawing of the needle-electrode systemplaced in the posterolateral spinal cord for an extralemniscal-myelotomy and pertinent anatomical structures.




lww/nq/88532/04-123

7. Kanpolat Y, Akyar S, Caglar S, et al. CT-guided percutaneous selectivecordotomy. Acta Neurochir (Wien). 1993;123:92–97.

8. Kanpolat Y, Akyar S, Caglar S. Diametral measurements of the upperspinal cord for stereotactic pain procedures: experimental and clinicalstudy. Surg Neurol. 1995;43:478–483.

9. Kanpolat Y, Savas A, Caglar S, et al. Computerized tomography-guidedpercutaneous bilateral selective cordotomy. Neurosurg Focus. 1997;4:1–5.

10. Kanpolat Y, Savasx A, Batay F, et al. Computed tomography-guidedtrigeminal tractotomy-nucleotomy in the management of vagoglossophar-yngeal and geniculate neuralgias. Neurosurgery. 1998;43:484–490.

11. Sweet WH, Mark VH. Unipolar anodal electrolytic lesion of man and cat:report of five human cases with electrically produced bulbar or mesen-cephalic tractotomies. Arch Neurol Psychiatry. 1953;70:349–350.

12. Cosman ER, Cosman BJ. Methods of making nervous system lesions. In:Wilkins RH, Renchary SS, eds. Neurosurgery. New York: McGraw-Hill;1985:2990–2999.

13. Cosman ER, Nashold BS, Bedenbaugh P. Stereotactic radiofrequencylesion making. Appl Neurophysiol. 1983;46:160–166.

14. Levin AB, Cosman ER. Thermocouple-monitored cordotomy electrode.J Neurosurg. 1980;53:266–268.

15. Rosomoff HL, Carroll F, Brown J, et al. Percutaneous radiofrequencycervical cordotomy: technique. J Neurosurg. 1965;23:639–644.

16. Crue BL, Carregal JA, Felsoory A. Percutaneous stereotactic radio-frequency. Trigeminal tractotomy with neurophysiological recordings.Confin Neurol. 1972;34:389–397.

17. Hitchcock E. Stereotactic cervical myelotomy. J Neurol NeurosurgPsychiatry. 1970;33:224–230.

18. Hitchcock ER. Stereotactic trigeminal tractotomy. Ann Clin Res. 1970;2:131–135.

19. Schvarcz JR. Stereotactic extralemniscal myelotomy. J Neurol NeurosurgPsychiatry. 1976;39:53–57.

20. Todd EM, Crue BL, Carregal JA. Posterior percutaneous tractotomy andcordotomy. Confin Neurol. 1969;31:106–115.

21. Kanpolat Y, Cosman E. Special RF electrode system for CT-guided painprocedures. Neurosurgery. 1996;38:600–603.

22. Everet NB, ed. Functional Neuroanatomy. Philadelphia: Lea & Febiger;1967:34–67.

23. Walker EA. The spinothalamic tract in man. Arch Neurol Psychiatry.1940;43:284–298.

24. Hyndman OR, Van Epps C. Possibility of differential section of thespinothalamic tract. A clinical and histologic study. Arch Surg. 1939;38:1036–1053.

25. Martin JH, Jessel TM. Anatomy of the somatic sensory system. In: KandelE, Schwartz J, Jessel TM, eds. Principles of Neural Science. New York:Elsevier; 1991:353–366.

26. Belmusto L, Brown E, Owens G. Clinical observations on respiratory andvasomotor disturbance as related to cervical cordotomies. J Neurosurg.1963;20:225–232.

27. Lorenz R. Methods of percutaneous spinothalamic tract section. In:Krayenbuhl H, ed. Advances and Technical Standards in Neurosurgery,vol. 3. Vienna: Springer Verlag; 1976:123–154.

28. Mullan S. Percutaneous cordotomy. J Neurosurg. 1971;35:360–366.29. Rosomoff HL, Krieger AJ, Kuperman AS. Effects of percutaneous cer-

vical cordotomy on pulmonary function. J Neurosurg. 1969;31:620–627.30. Rosomoff HL. Bilateral percutaneous cervical radiofrequency cordotomy.

J Neurosurg. 1969;31:41–46.

31. Sindou M, Jeanmonod D, Mertens P. Ablative neurosurgical proceduresfor the treatment of chronic pain. Neurophysiol Clin. 1990;20:399–423.

32. Tasker RR. Percutaneous cordotomy for persistent pain. In: GildenbergPL, Tasker RR, eds. Textbook of Stereotactic and Functional Neurosur-gery. New York: McGraw-Hill; 1998:1491–1505.

33. Brihaye J, Thiry S, Le Clerco R, et al. Le traitement chirurgical de ladouleur. Acta Chir Belg. 1962;2(suppl):255–275.

34. Peet MM. The control of intractable pain in lumbar region, pelvis andlower extremities. Arch Surg. 1926;13:153–204.

35. Kanpolat Y, Deda H, Akyar S, et al. CT-guided percutaneous cordotomy.Acta Neurochir Suppl (Wien). 1989;46:67–68.

36. Kanpolat Y, Deda H, Akyar S, et al. CT-guided trigeminal tractotomy.Acta Neurochir (Wien). 1989;100:112–114.

37. Hosobuchi Y, Rutkin B. Descending trigeminal tractotomy. Neurophys-iological approach. Arch Neurol. 1971;25:115–125.

38. Schvarcz JR. Postherpetic craniofacial dysesthesiae: their managementby stereotactic trigeminal nucleotomy. Acta Neurochir (Wien). 1977;38:65–77.

39. Schvarcz JR. Descending trigeminal nucleotomy for dysesthetic facialpain. In: Al-Mefty O, Origitano TC, Haarkey HL, eds. Controversies inNeurosurgery. Stuttgart: Georg Thieme Verlag; 1996:351–353.

40. Sweet WH, Poletti CE. Operations in the brain stem and spinal canal, withan appendix on open cordotomy. In: Wall PD, Melzack R, eds. Textbook ofPain. 2nd ed. Edinburgh: Churchill Livingstone; 1989:819–820.

41. White JC, Sweet WH. Spinothalamic tractotomy. In: Pain and theNeurosurgeon. Springfield, IL: Charles C Thomas; 1969:678–772.

42. Nashold BS, Rossitch E. Anesthesia dolorosa and the trigeminal caudalisnucleus DREZ operation. In: Rovit RL, Murali R, Janetta PJ, eds.Trigeminal Neuralgia. Baltimore: Williams & Wilkins; 1990:223–237.

43. Taren JA, Kahn EA. Anatomic pathways related to pain in face and neck.J Neurosurg. 1962;19:116–121.

44. Burchiel K. Treatment of facial pain. In: Gildenberg PL, Tasker RR, eds.Textbook of Stereotactic and Functional Neurosurgery. New York:McGraw-Hill; 1998:1723–1728.

45. Kerr LWF. The fine structure of the subnucleus caudalis of the trigeminalnerve. Brain Res. 1970;23:29–145.

46. Sjoqvist O. Studies on pain conduction in the trigeminal nerve. Acontribution to the surgical treatment of facial pain. Acta Psychiatr NeurolScand. 1938;(suppl 17):93–122.

47. Teixeira MJ. Various functional procedures for pain. Part II: facial pain. In:Gildenberg PL, Tasker RR, eds. Textbook of Stereotactic and FunctionalNeurosurgery. New York: McGraw-Hill; 1998:1389–1402.

48. Nashold BS, El-Naggar A, Abdulhak MM, et al. Trigeminal nucleuscaudalis dorsal root entry zone: a new surgical approach. Stereotact FunctNeurosurg. 1992;59:45–51.

49. Kanpolat Y, Atalag M, Deda H, et al. CT-guided extralemniscalmyelotomy. Acta Neurochir (Wien). 1988;91:151–152.

50. Kanpolat Y, Savasx A, Cxaglar Sx, et al. Computerized tomography-guidedpercutaneous extralemniscal myelotomy. Neurosurg Focus. 1997;2:1–4.

51. Gildenberg PL, Hirshberg RM. Treatment of cancer pain with limitedmyelotomy. Med J St Jos Hosp. 1981;16:199–204.

52. American Cancer Society. Cancer Statistics CA. 1997;47:8–12.53. Foley KM. The treatment of cancer pain. N Engl J Med. 1985;311:

84–95.54. World Health Organization. Cancer as a global problem. Epidemiol Rec.

1984;59:125–126.




lww/nq/88532/04-123

Documents

Percutaneous Computed Tomography-Guided … Computed Tomography-Guided Spinal Destructive Procedures for Pain ... he spinal cord, ... surgery had to be performed under general anesthesia,