PAIN MEDICINEVolume 7 • Number 5 • 2006

© American Academy of Pain Medicine 1526-2375/06/$15.00/424 424–428

Blackwell Publishing IncMalden, USAPMEPain Medicine1526-2375American Academy of Pain Medicine200675424428Original ArticlePulsed Radiofrequency for Discogenic PainTeixeira and Sluijter

Reprint requests to: Alexandre Teixeira, MD, Clinica de Dor,Rua Sao Joao Brito 610, 1°, Sala 7, 4100-453 Porto, Por-tugal. Tel: +351-615-1367; Fax: +351-615-1368; E-mail:alteix@iol.pt.

Preliminary Research

Intradiscal High-Voltage, Long-Duration Pulsed Radiofrequency for Discogenic Pain: A Preliminary Report

Alexandre Teixeira, MD,* and Menno E. Sluijter, MD, PhD†

A B S T R A C T

*Clinica de Dor, Porto, Portugal; †Pain Unit, Swiss Paraplegic Center, Nottwil, Switzerland

ABSTRACT Background. Intradiscal radiofrequency, with the electrode placed in the center of the nucleuspulposus, has been a controversial procedure in patients with discogenic pain. Possibly the effecthas not been due to the production of heat, but to exposure to electric fields.

Design. We have investigated the effect of high-voltage, long-duration intradiscal pulsed radio-frequency in patients with one-level discogenic low back pain, as confirmed by discography.

Outcome Measure. The pain intensity score on a 0–10 numeric rating scale (NRS) was taken asoutcome measure.

Patients. Eight patients were reported. The mean duration of pain was 6.3 years (range 0.5–16,median 4). The mean NRS score was 7.75 (range 5–9). Disc height was reduced 60% in one patientand up to 30% in the others.

Intervention. A 15-cm, 20-gauge needle with a 15-mm active tip was placed centrally in the disc.Pulsed radiofrequency was applied for 20 min at a setting of 2 × 20 ms/s and 60 V.

Results. There was a very significant fall in the NRS scores over the first 3 months (P < 0.0001).On an individual basis, all patients had a fall of the NRS score of at least 4 points at the 3-monthfollow-up. A follow-up of 12.8 months (range 6–25, median 9) was available for five patients. Allthese patients are now pain free, except for one patient with an NRS score of 2.

Conclusion. It is concluded that this method merits a controlled, prospective study.

Key Words. Pulsed Radiofrequency; Discogenic Pain; Intradiscal Procedures

Introduction

ntradiscal radiofrequency (RF) was introducedin 1996 [1] as a form of treatment for discogenic

pain. It was based on the principles that the ver-tebral endplates serve as heat insulators, and thatthe center of the disc is avascular. These factors

Itogether lead to an accumulation of heat inside thedisc, potentially affecting the nerve endings in theannulus fibrosus.

Since the introduction of pulsed RF (PRF) [2],the role of heat in producing the clinical effect ofRF lesions has become controversial. If the effectof intradiscal RF has in fact been due to exposureto RF electric fields, instead of to the formation ofheat, the technique may have been less than opti-mal. The duration of the heat lesion was relativelyshort, and the voltage was limited, as it wasadjusted to the production of heat.

Pulsed Radiofrequency for Discogenic Pain 425

We have investigated the effect of intradiscalPRF using higher voltage and longer exposuretime. We here report about the first eight patientswho have been treated with this novel technique.

Patient Material

The study includes eight consecutive patients withdiscogenic pain, who presented at the Clinica deDor, Porto, Portugal. The demographic data areshown in Table 1. All patients had had continuousback pain without referral to the legs for a mini-mum of 6 months. The mean duration of pain was6.3 years (range 0.5–16, median 4). The meanscore on a 0–10 numeric rating scale (NRS) was7.75 (range 5–9). Seven patients took a variety ofmedication, including various nonsteroidal anti-inflammatory drugs (NSAIDS), cyclooxygenase(COX)-inhibitors, diazepam for muscle relaxation,amitryptyline, and tramadol hydrochloride. Onepatient had stopped all medication, because it didnot provide any relief of pain. Five out of sixpatients having a profession had stopped workingbecause of their back pain.

All patients had normal findings on neurologi-cal examination, including a negative Straight LegRaising test. On radiological examination, narrow-ing of the height of the affected disc of up to 30%was found in seven patients. In one patient, therewas 60% narrowing of the disc space. Two patientshad previously undergone disc surgery for a her-niated disc. In one of them, the pain originatedfrom the operated level; in the other from the levelcranial to the level of operation.

In all patients, a diagnostic medial branch blockat the relevant segments had excluded a diagnosisof facet pain. The diagnosis of single-level disco-genic pain had been confirmed by discography atminimally two levels. Provocation of concordantpain was reported at manually controlled low-pressure injection of contrast medium, and in allpatients, the discogram was abnormal.

All patients had been informed of the experi-mental nature of the proposed treatment, andwritten consent was obtained.

Technique

Patients were treated on an outpatient basis. Seda-tion was not used.

Sterile conditions were maintained throughoutthe procedure. Using an oblique projection and atunnel vision technique [3], a 20-gauge SMK C15canula with a 15-mm active tip (Cotop Interna-tional, Amsterdam, The Netherlands) was aimedat the center of the affected disc. It was then slowlyintroduced further, and special care was taken toavoid contact with the exiting segmental nerve.Once the disc had been entered, the canula wasfurther advanced until the tip was just past thecenter of the disc.

The stylet was then removed and replacedby the RF probe. The mean impedance was140.6 ± 11.4 Ohms. PRF was now applied at asetting of 2 × 20 ms/s and 60 V, for a duration of20 min. The mean current intensity during theactive phase was 520.9 ± 27.6 mA, and the meanfinal tip temperature was 41.0 ± 2.0°C.

At the end of the procedure, 100 mg ofCefamezin (Hikma Farmaceutica Ida, Lisbon,Portugal) was injected into the disc to prevent theoccurrence of a discitis.

Patients were instructed to avoid heavy workand strenuous exercise, and to stop physiotherapyfor a period of 2 weeks. Otherwise, no restrictionof activity was prescribed.

Results

Clinical Course During the First WeekDaily NRS scores over this period are available forsix patients. There was a very significant fall inNRS score (P < 0.0001, Friedman test, nonpara-metric repeated measures anova, two-tailed). Thedata are shown in Figure 1.

Clinical Course During the First 3 MonthsThere was a very significant fall in the NRS scoresover the first 3 months (P < 0.0001, Friedmantest, nonparametric repeated measures anova,two-tailed). The data are shown in Figure 2. Onan individual basis, all patients had a fall of theNRS score of at least 4 points at the 3-monthfollow-up.

Long-Term Follow-upA follow-up of 12.8 months (range 6–25, median9) is available for five patients. All these patientsare now pain free, except for one patient with an

Table 1 Demographic data

N Mean Age Range

Male 5 44.0 39–48Female 3 60.0 34–75

Total 8 50.0 34–75

426 Teixeira and Sluijter

NRS score of 2. In none of the patients has therebeen a recurrence of pain.

Professional Activity and MedicationOut of the five patients who had stopped working,one patient lost her job due to frequent absentee-ism during the pretreatment period. The otherpatients went back to work after a mean period of34.5 days (range 7–90).

All patients stopped their medication, exceptfor an occasional NSAID.

Discussion

The present study is not based on any new theoryregarding PRF. It was initiated to explore the pos-sibility that the results of intradiscal heating havebeen due to exposure to electric fields rather thanto the generation of heat. Intradiscal RF, with theelectrode positioned in the middle of the disc, hasnot been a successful procedure. It was based onthe idea that the heat would reach the nerve end-ings in the annulus fibrosus, because the disc spacelays between the vertebral endplates that functionas a thermal insulator. The heat would thereforebe forced to expand sideward. The initial publica-tion [1] reported acceptable results, but a random-ized controlled trail could not confirm clinicaleffectiveness [4].

The failure of the method may in retrospect bedue to one of two reasons. First, heating of theannulus may have been ineffective [5]. Second,the mode of action may not have been due to thedevelopment of heat at all. The concept of the

method dates back to before the discussion onthe role of RF-generated heat in pain treatment.If the clinical effect that was observed has been dueto exposure to RF electric fields, and not to theformation of heat, then the parameters of intradis-cal RF were probably inadequate, both in durationand in intensity of exposure.

The mode of action of intradiscal PRF isunknown. It has nothing to do with heat, as thereis no appreciable heat production around the elec-trode tip. As far as we can see, there are two pos-sible hypotheses. First, the combination of thehigh voltage and the low impedance inside thenucleus pulposus must cause very strong electricfields during the active phase of the duty cycle, andthese could potentially have a biological effect onthe nerve endings in the annulus fibrosus.

Second, there may be an effect on the compo-sition of the nucleus pulposus. In a nociceptivedisc, the pH is lower than normal [6], and there isan increased content of phospholipase A [7]. Thisaltered composition might be responsible for trig-gering nerve endings in patients with discogenicpain, both in the annulus fibrosus and more cen-trally in a degenerative disc [8]. If the electric fieldsaffect the composition of the nucleus, PRF wouldwork by modifying the source of nociception,rather than by ablating or modifying a signal. Thiswould be a new concept.

Another novel aspect of our method is the useof a high voltage and of a long exposure time forPRF. We feel that this is an essential part of thetechnique for this particular procedure. For otherindications, a voltage of 45 V and an exposure time

Figure 2 Mean NRS scores and standard deviationsduring the first 3 months after treatment (N = 8).NRS = numeric rating scale.

InitialMonths

Figure 1 Mean NRS scores and standard deviations dur-ing the first week after treatment (N = 6). NRS = numericrating scale.

Days

Initial

Pulsed Radiofrequency for Discogenic Pain 427

of 2 min have been customarily used for PRFtreatment. These values are arbitrary [3], andthere are no studies available using different volt-ages and exposure times. It is too early to tell ifthe success of longer exposure times in this smallinitial study, on this particular application, willnecessitate reappraisal of the parameters in otherPRF procedures.

The use of the needle with a very long activetip may have contributed to the effect as well. Ithas initially been thought that the electric fieldduring the active pulse would mainly be distrib-uted forward, ahead of the needle tip [3]. Com-puter calculations of electric fields are indeedfound ahead of the tip, but they wear off rapidlyover a very short distance of 0.1–0.2 mm. At alarger distance, the magnitude of the field iseven somewhat larger in the area that is parallelto the uninsulated part of the needle. Contraryto what was thought, the length of the active tipis therefore an important element of a PRFtechnique [9].

A radiographic follow-up has not been con-ducted in this study because even a thermal lesioninside the nucleus does not cause any magneticresonance imaging (MRI) changes [1]. As there isvery little heat formation in the proceduredescribed here, it is unlikely that any changeswould have been found.

We find the results of this study very satisfac-tory, but nevertheless the question should beanswered if there is a need for a new method totreat discogenic pain. PRF treatment of the L2dorsal root ganglion [3,10] is an elegant, well-tolerated method, but it has a success rate notexceeding 50%. Other methods for heating theannulus fibrosus are now available. The mode ofaction of these methods is not known. Theyintend to destroy the nerve endings in the annu-lus fibrosus, but this does not actually happen[11].

The initial publication about intradiscal elec-trothermal anuloplasty (IDET) [12] reported agood result in over 70% of patients. Later reports[13–16], including a randomized study [17], wereless optimistic, indicating that the method is anexcellent form of treatment in a small proportionof patients, but that on the other hand an increasein pain is not uncommon. Also the method is con-traindicated for technical reasons if there is a lossin disc height of >50%, and the method is not freeof complications [18,19].

Another instrument for heating the annulusfibrosus is the DiscTRODE (Radionics, Boulder,

CO). In the first report on the outcome of thismethod [20], a very significant fall in the meanvisual analog scale (VAS) score was found, but asin IDET this was due to a very favorable outcomein a limited number of patients. Less than 30%improvement in the VAS score was reported in52% of treated patients.

We conclude that the ideal method to treatdiscogenic pain has not yet been found. Thereseems to be a good reason to conduct a pro-spective, controlled study to validate this newmethod.

References

1 Van Kleef M, Barendse GAM, Wilmink JT, et al.Percutaneous intradiscal radiofrequency thermoco-agulation in chronic non-specific low back pain.Pain Clin 1996;9:259–68.

2 Sluijter ME, Cosman E, Rittman W, van Kleef M.The effect of pulsed radiofrequency fields applied tothe dorsal root ganglion—A preliminary report.Pain Clin 1998;11:109–17.

3 Sluijter ME. Radiofrequency, Part 1. Meggen, Swit-zerland: Flivopress; 2001.

4 Barendse GA, van den Berg SG, Kessels AH, WeberWE, van Kleef M. Randomized controlled trial ofpercutaneous intradiscal radiofrequency thermoco-agulation for chronic discogenic back pain: Lack ofeffect from a 90-second 70 c lesion. Spine 2001;26:287–92.

5 Troussier B, Lebas JF, Chirossel JP, et al. Percuta-neous intradiscal radio-frequency thermocoagula-tion. A cadaveric study. Spine 1995;20(15):1713–8.Comment in: Spine February 15, 1996;21(4):528–9.

6 Kitano T, Zerwekh JE, Usui Y, et al. Biochemicalchanges associated with the symptomatic humanintervertebral disk. Clin Orthop Relat R 1993;293:372–7.

7 Piperno M, le Graverand MPH, Reboul P, et al.Phospholipase A2 activity in herniated lumbar discs.Spine 1997;22:2061–5.

8 Coppes MH, Marani E, Thomeer RT, Groen GJ.Innervation of “painful” lumbar discs. Spine 1997;22:2342–9.

9 Cosman ER, Cosman ER. Electric and thermal fieldeffects in tissue around radiofrequency electrodes.Pain Med 2005;6(6):405–24.

10 Gauci CA. Pulsed radiofrequency for the manage-ment of discogenic low back pain abstract. WorldWide Pain Conference, San Francisco, 2000.

11 Freeman BJ, Walters RM, Moore RJ, Fraser RD.Does intradiscal electrothermal therapy denervateand repair experimentally induced posterolateralannular tears in an animal model? Spine 2003;28(23):2602–8.

428 Teixeira and Sluijter

12 Saal JA, Saal JS. Intradiscal electrothermal treat-ment for chronic discogenic low back pain: Aprospective outcome study with minimum 1-yearfollow-up. Spine 2000;25:2622–7.

13 Spruit M, Jacobs WC. Pain and function after intra-discal electrothermal treatment (IDET) for symp-tomatic lumbar disc degeneration. Eur Spine J2002;11(6):589–93.

14 Freedman BA, Cohen SP, Kuklo TR, et al. Intra-discal electrothermal therapy (IDET) for chroniclow back pain in active-duty soldiers: 2-year follow-up. Spine J 2003;3(6):502–9.

15 Davis TT, Delamarter RB, Sra P, Goldstein TB.The IDET procedure for chronic discogenic lowback pain. Spine 2004;29(7):752–6.

16 Karasek M, Bogduk N. Twelve-month follow-up ofa controlled trial of intradiscal thermal anuloplasty

for back pain due to internal disc disruption. Spine2000;25:2601–7.

17 Pauza KJ, Howell S, Dreyfuss P, et al. A random-ized, placebo-controlled trial of intradiscal electro-thermal therapy for the treatment of discogenic lowback pain. Spine J 2004;4(1):27–35.

18 Hsia AW, Isaac K, Katz JS. Cauda equina syndromefrom intradiscal electrothermal therapy. Neurology2000;55(2):320.

19 Djurasovic M, Glassman SD, Dimar JR II, JohnsonJR. Vertebral osteonecrosis associated with the useof intradiscal electrothermal therapy: A case report.Spine 2002;27(13):E325–8.

20 Finch PM, Price LM, Drummond PD. Radiofre-quency heating of painful annular disruptions: One-year outcomes. J Spinal Disord Tech 2005;18(1):6–13.