:Acta . Ndurochlrurgica

Acta Neurochir (Wien) (1990) 107:147-151

9 by Springer-Verlag 1990

Spinal Cord Stimulation in 112 Patients with Epi-/Intradural Fibrosis Following Operation for Lumbar Disc Herniation Ch. Probst Neurosurgical Clinic, Kantonsspital, Aarau, Switzerland

Summary A total of 112 patients with epi-/intradural fibrosis following operation for lumbar disc herniation were treated by spinal cord stimulation. Lumbosacral spinal fibrosis is seen particularly often after extensive and repeated operations. Radicular pain responds better to stimulation than back pain. A favourable long-term effect on radicular pain has been observed in 67% of patients treated by epidural implantation, the corresponding average follow-up period being 4 V~years. 40% of these patients needed less analgesics after the operation, while 25% of them showed an improved fitness for work. Among about 5,000 patients who underwent surgical treatment for lumbar disc herniation, an indication for spinal cord stimulation was found in 1,5%. By comparison, the frequency of the "last resort" procedure of microsurgical cordotomy was 0.3%. We no longer use other ablative methods like extirpation of spinal ganglia.

Keywords: Lumbosacral fibrosis; failed back surgery syndrome; lumbar disc surgery; spinal cord stimulation.

Introduction S e v e r e d i s a b l i n g l u m b a r a n d r a d i c u l a r p a i n c a n be c a u s e d by e p i - / i n t r a d u r a l fibrosis s u b s e q u e n t to o p e r a t i o n f o r l u m b a r disc h e r n i a t i o n . S p i n a l c o r d stimu l a t o r s w e r e d e f i n i t i v e l y i m p l a n t e d in 112 s u c h p a t i e n t s w i t h i n t r a c t a b l e pain. T h e i n d i c a t i o n s f o r this p r o c e d u r e , t o g e t h e r w i t h the o p e r a t i v e t e c h n i q u e a n d results are discussed. F i n a l l y , t h e p r o b l e m o f p o s t o p e r a t i v e e p i - / i n t r a d u r a l f i b r o s i s is c o n s i d e r e d f r o m the p a t h o genic, d i a g n o s t i c a n d t h e r a p e u t i c s t a n d p o i n t s , o n the basis o f m o r e t h a n 5,000 p a t i e n t s w h o h a v e u n d e r g o n e o p e r a t i o n f o r l u m b a r disc h e r n i a t i o n in o u r clinic.

In 114 patients the primary LDH operation was performed in our clinic, in 26 patients in other clinics, thus accounting for a frequency of secondary surgery for pain relief of 2%. At the same time the frequency of local re-operations for recurrent disc herniation at the same site or at another level or for a narrow recessus lateralis was 5.5%. During the period 1975- 1982, after testing with subarachnoid electrodes, endoduralstimulators were implanted in 20patients at the Th2 level (small laminectomy). Since 1982 we have carried out the test with epidural electrodes, percutaneously introduced under local anaesthesia, usually at the T h S - 10 level. When the result of the test (24hours) was favourable (completely free of radicular pain during stimulation or residual pain less than 50%), the same electrode was used as a definitive implant in 92patients. Up to the end of 1982, subcutaneously sited receivers were implanted in all cases, the amplitude and frequency being determined by the patient himself. Nowadays, the majority of patients receive totally implantable, programmable bipolar systems with intermittent stimulation, whereby epidurai electrodes are often introduced by puncture. In cases of bilateral pain, an epidurally located plate electrode introduced after a small laminectomy may also be used. Diagnosis of pain due to epio/intradural fibrosis was based on clinical and neuroradiological findings described in detail in the discussion. Among our 112 patients there were no signs of important spinal instability. In cases of doubt, a trial plaster jacket was used for three weeks. In two thirds of the 112 cases the radicular leg pain was unilateral, in one third bilateral. Very often radicular pain was slowly increasing and combined with polysegmental sensory disturbances. Mostly there was also low back pain, and in 6 patients, this was the predominant symptom. 85% of our patients had a history of extensive (exploration at various levels or bilaterally) and repeated operations. The group of ll2patients studied consists of 64men and 48 women. Their mean age is 48 years, the duration of pain prior to the implantation 1 - 33 years, the postoperative follow-up period 6-168 months (mean 54 months) (Table 2).

Patients and Method During the period 1973- 1987, a total of about 5,000 patients underwent operation for lumbar disc herniation (LDH). During the same period, 140patients were treated for recurrent or persistent radicular and lumbar pain due to postoperative epi-/intradural fibrosis (Table 1).

Results Effect on Pain T h e best results w e r e o b t a i n e d f o r t h e radicularpain. T h e s e results are s u m m a r i z e d in T a b l e 3. L o n g - t e r m

148

Ch. Probst: Epi-/Intradural Fibrosis Following Operation for Lumbar Disc Herniation

Table 1. Operations for Relief of Pain due to Symptomatic Epi-/ Intradural Fibrosis After LDH Operations LDH operations Surgery for pain relief Spinal cord stimulation (since 1975) Microsurgical cordotomy (since 1973) Exstirpation of spinal ganglia (1973-1976)

5,000 pat. ll4pat. (140)* 86 pat. (112)* 20 pat. (20)* 8 pat. (8)*

Table 2. Spinal Cord Stimulation in 112 Patients with Postoperative Lumbar Epi-/Intradural Fibrosis Sex distribution Age Duration of pain before stimulation Follow-up period after implantation

* Total: including patients with a primary LDH operation in other clinics.

64 men, 48 women 30-76 years (mean 48 years) 1-33 years (mean 4 years) 6-168 months (mean 54 months)

Table 3. Results Concerning Radicular Pain Postop.

6-12 months

13 24 months

longterm

50% (10) 15% (3) 15% (3) 20% (4)

30% (6) 15% (3) 15% (3) 40% (8)

30% (6) 15% (3) 15% (3) 40% (8)

44% (40) 33% (31) 15% (14) 8% (7)

38% (34) 31% (29) 16% (15) 15% (14)

36% (33) 31% (29) 18% (16) 15% (14)

a) Endodural Stimulators, 20 pat. (1975-1982) I. II. III. IV.

Pain-free Residual pain less than 50% Residual pain more than 50% No effect on pain

66% (14) 33% (6)

b) Epidural Stimulators, 92 pat. (1982-1987) I. II. III. IV.

Pain-free Residual pain less than 50% Residual pain more than 50% No effect on pain

71% (66) 29% (26)

results concerning radicular pain obtained by epidural implantations are better (67% with very good and good results, respectively) than they are after endodural operations (45%). The effect on back pain was less marked: 25% of the patients showed very good or good results. Here the results are, however, occasionally so good, that a test can be thoroughly recommended. All the patients had good results immediately after the operation, corresponding to the test results. The recurrence rate was highest in the first postoperative year, followed by the second year. The majority of recurrences occurred during these first two years. Relapses were also recorded in the presence of a properly functioning stimulator - for no obvious reason (21 patients). In 6 patients, progressive degenerative changes with additional root compression were later observed, while 4 patients became severely dependent on analgesics. 40 % of patients treated by stimulation subsequently needed no further analgesics, or their dependence on them was considerably reduced. In 4 patients stimulation abolished a labile hypertension, which was obviously increased by the pain, and so, these patients

were able to dispense with antihypertensive medication after the implantation. Fitness for work was improved in 25% of patients, whereby all of them were disabled before operation. Complications, Reoperations Severe neurological or other complications did not occur. In 32 of 92 patients (28%) with epidural implantation, one to five reoperations became necessary for the following reasons: displacement or instability of electrodes (20 patients), technical faults (7 patients), infection (5 patients). In two of the 5 patients with local infection, the system could be retained after local debridement and suction/irrigation drainage with prim a r y wound closure. N o n e of the 20 patients with endodural electrodes was reoperated upon.

Discussion Epi-/Intradural Fibrosis Following Operation for Lumbar Disc Herniation." Causes, Diagnoses The report of Spiller et al. 24 in 1903 constitutes the first mention of so-called "arachnoiditis" as a patho-

Ch. Probst: Epi-/Intradural Fibrosis FollowingOperation for Lumbar Disc Herniation logical condition in the spinal region. Attention was drawn by Smolik et al. 23 in 1951 to the aetiological connection between this condition and the operative treatment of disc herniation. Epidural/intradural fibrosis may remain clinically asymptomatic4. When it becomes the cause of symptoms, in most cases slowly increasing uni- or bilateral radicular pain combined with polysegmental sensory disturbances are observed, rarely a low back syndrome is predominant. The neuroradiologicalfindings are important for the differential diagnosis. A positive enhancement in the CT following contrast administration is typical of epidural fibrosis, in contrast to herniation of the nucleus pulposus 17. Occasionally, epidural fibrosis and a small recurrent disc herniation may be observed simultaneously. Intradural fibrosis is visible in CT only when it is calcified. The myelogram may reveal chronic lumbar arachnoiditis including defective or absent root sheaths up to shortening of lumbar dural sac or a total stop. In doubtful cases, it is advisible to perform a functional myelogram, in order to reveal a possible root compression, which becomes visible only during sitting or standing. Similar symptoms may be observed in the case of an unstable segment following lumbar disc surgery. A trial with a plaster jacket - for external stabilization - will lead to complete or nearly complete disappearance of the symptoms in such cases 15' 16 Lumbosacral spinal fibrosis may have various causes: Epidurally it may follow operations for lumbar disc herniation, where local haemorrhages, the postoperative invasion of necrotic muscle tissue into the epidural space and infections all may play a role 12 14, as does also absence of the epidural fatty tissue 1~ 11 With the aid of optic magnification it is possible to perform very sparing explorations. In patients with monosegmental disc herniation a small incision of 3 cm length is sufficient 14. The more extensive and more frequent the local exploration, the poorer is the final result. This applies to postoperative vertebral symptoms I4 and also to radicular signs. 85% of our patients admitted for pain surgery following LDH-operations had a complex history of repeated and/or extensive explorations on both sides or at several levels. Intradural fibrosis (often referred as "spinal arachnoiditis") may be the result of severe root compression, especially when this is of chronic type. It may be caused by diffuse or circumscribed degenerative narrowing of the lumbar canal 5, and it also observed after myelography. The number of myelograms and the choice of contrast medium I' 6, 7. 22 certainly is of importance.

149

Overproduction of scar tissue is also seen in unstable segments after disc excision and partial or complete facetectomy. Operative Treatment It is widely accepted that results of local reoperation (neurolysis) for symptomatic epidural fibrosis often are disappointing. Benoist et al. 2 reported that in 13 of 38 patients local re-operation was followed by decrease of radicular pain, while 8 showed partial and 17 showed no improvement. Johnston et al. s carried out local reoperations in 27 patients and found it to be useless compared with conservative treatment. Wilkinson et aI. 26 observed a postoperative improvement of pain in about half of their group of 17 patients, neurologically 18 % of them were in worse condition. According to Thomalske et a l Y , the chances of a successful outcome are only 37%. Brodsky 3, however, had a relatively high proportion of 64% good results. It is our policy to carry out local re-operation only in patients with new signs of root compression due to a new disc herniation at the previously operated or another site and/or with narrowing of the recessus lateralis. Among the 5,000 patients who underwent lumbar disc-operations during the period 1973-1987, the frequency of local re-operation was 5.5%. It is important to exclude an unstable segment as the main cause of the symptoms, which may be superimposed by secondary scar formation. In such cases, one possible treatment is decompression of the affected root, followed by internal fixation. Markwalder and Reulen have recently published their results with this techniquelS, 16 In cases of postoperative epi-intradural lumbosacral fibrosis without important signs of instability, severe intractable radicular and/or lumbar pain may be an indication for spinal cord stimulation, first described by Shealy et alJ 9 in 1967. This method has proved effective in cases of deafferentation-pain: Phantom pain, peripheral neural lesions with causalgia, plexus fibrosis following irradiation and lumbosacral fibrosis. Since 1982, we have implanted epidural electrodes. The method is simple and sparing for the patient. Since August 1985 we have preferred totally implantable bipolar systems programmed exteriorly. According to Shatin et al.lS, the intermittent stimulation gives more promising results than the continuous one. Severe complication (death, neurological deficits) do not occur. Although local re-operation for a variety of reasons is quite common (in 40% of the patients 4 and in 28% of our own cases), this can usually be carried out under

150

Ch. Probst: Epi-/Intradural Fibrosis Following Operation for Lumbar Disc Herniation

local anaesthesia. Most important improvements are the use of multicontact electrodes and the possibility of transcutaneous change of polarity/level of stimulation, modulation of PW and frequency, such as amplitude switch. These possibilities will partly avoid reoperations for electrode replacement and will be advantageous in cases, where the stimulation effect depends on position. In cases of local infection, the system can often be retained after local debridement followed by suction/irrigation drainage with primary wound closure. With a follow-up period of 4 89 years, the long-term results as regards radicular pain were good in 67% of epidural stimulation cases 4' 9, 13, 20, 21. Relapse of the pain occurs mainly during the first postoperative year. Radicular pain responds better than back pain, although we have had very good long-term results also in a number of patients who presented with an isolated lumbovertebral syndrome. It is interesting to note that better long-term results are obtained after epidural implantation (67%) than after the endodural implantation (45%). As regards postoperative reduction in the consumption of analgesics (40%) and improvement in fitness for work (25%), our experiences agree with those of other authors H. In our clinic, spinal cord stimulation is today the preferred method of treating very severe and intractable lumbar/radicular pain due to epi-/intradural fibrosis following LDH-operations. When this fails to help, microsurgical cordotomy can be used as a "last resort". According to our experience, the improved cordotomytechnique yields long-term results concerning radicular pain of about the same order as spinal cord stimulation. Although we did not see serious pyramidal tract lesions after cordotomy in patients with benign lesions 13, the risk still exists. In contrast, spinal cord stimulation carries no severe risk. We have abandoned completely extirpation of spinal ganglia, because good long-term results were obtained in only a quarter of our patients 13. Removal of the efferent cells of the sensory fibres in the spinal ganglion also interrupts sensory impulses conveyed via the anterior nerve roots. The problem of pain conduction via the sympathetic trunk ( T h l 0 - 12 for the leg) persists as well as that of segmental overlapping: an excessively extensive and radical interruption of the sensory pathways of the extremities is impossible since deep sensibility must be maintained 13 One problem of destructive pain-surgery remains, namely the possiblity of postoperative deafferentation pain. Therefore, we prefer neuro-stimulation in all cases where this is possible. Spinal opioid administration by

implanted infusion pumps is another possibility. Our first observations in such cases are very encouraging. References

1. Ahlgren P (1978) Amipaque myelographywithout late adhesive arachnoid changes. Ncuroradiology 14:231-233 2. Benoist M, Ficat C, Baraf Pet al (1979) Sciatiques postop~ratoires par fibrose 6pidurale et arachnoidite lombaire. Rev Rhum 46/11:593-599 3. Brodsky AE (1978) Cauda equina arachnoiditis. A correlative clinical and roentgenologic study. Spine 3/1:51-60 4. De la Porte Ch, Siegfried J (1983) Lumbosaeral spinal fibrosis (spinal arachnoiditis). Its diagnosis and treatment by spinal cord stimulation. Spine 8/6:593-603 5. Epstein JA, Epstein BS, Lavine LS, Rosenthal AD, Decker RE, Carras R (1978) Obliterativearachnoiditis complicatinglumbar spinal stenosis. J Neurosurg 48:252-258 6. Hansen E, Fahrenkrug A, Praestholm J (1978) Late meningeal effects of myelographiccontrast media with special reference to metrizamide. Br J Rad 51:321-327 7. Haugthon V, Khang-Cheng HO (1980) The risk of arachnoiditis from experimental nonionic contrast media. Radiology 136: 395-397 8. Johnston JDH, Matheny JB (1978) Microscopiclysis of lumbar adhesive arachnoiditis. Spine 3/1:36-39 9. Krainick JU, Lazorthes Y, Probst Ch, Siegfried J, Steude U, Thoden U, WinkelmiillerW (1979) Long-term follow-upof dorsal column stimulation for pain of some European clinics. Joint study of the Dept of Neurosurgery of Freiburg (West-Germany), Toulouse (France), Aarau (Switzerland), Ziirich (Switzerland), Munich (West-Germany),Hannover (West-Germany).Ii dolore 1/2:91-95 10. Langenski61dA, Kiviluoto O (1976) Prevention of epidural scar formation after operations on the lumbar spine by means of free fat transplants. A preliminary report. Clin Ortbop 115:92-95 11. Po-Quang C, Chyun-Yu YY (1985) Prevention of postlaminectomy membrane - experimental and clinical observation. 4. Internationale Tagung des Weltverbandesder Wirbelsfiulenchirurgen, Wien (Abstract) 12. Probst Ch (1986) Mikrochirurgie bei lumbalen Diskushernien: Erfahrungen bei 4,000 operierten Patienten. Praxis 24:719-724 13. Probst Ch (1987) Persistierend-rezidivierendeLumboischialgien nach operierten Diskushernien: Schmerzchirurgische M6glichkeiten. Akt Rheumatol 12:157-160 14. Probst Ch (1989) Lumbale Diskushernien (5,000 operierteF~lle), Mikrochirurgieja oder nein. Neurochirurgia (Stuttg) 6/32: 172176 15. Markwalder Th M, Reulen HJ (1989) Diagnostic approach in instability and irritative state of a "lumbar motion segment" following disc surgery - failed back surgery syndrome. Acta Neurochir (Wien) 99:51-57 16. Markwalder Th M, Reulen HJ (1989) Translaminar screw fixation in lumbar spine pathology. Technicalnote. Acta Neurochir (Wien) 99:58 60 17. SchubigerO (1984) Die Computertomographieder Wirbcls/iule. In: Schulitz KP (Hrsg) Die Wirbels/iulein Forschung und Praxis, Band 104. Hippokrates 18. Shatin D, Mullett K, Hults G (1986) Totally implantable spinal cord stimulation for chronic pain: Design and efficacy.Pace 9: 577-583

Ch. Probst: Epi-/Intradural Fibrosis Following Operation for Lumbar Disc Herniation 19. Shealy C, Mortimer J, Reswick J (1967) Electrical inhibition of pain by stimulation of the dorsal columns: A preliminary report. Anesth Analg 46:489-491 20. Siegfried J, Cetinalp E (1981) Neurosurgical treatment of phantom limb pain: A survey of methods. In: Siegfried J, Zimmermann M (eds) Phantom and stump pain. Springer, Berlin Heidelberg New York, pp 148-155 21. Siegfried J, Lazorthes Y (1982) Long-term follow-up of dorsal cord stimulation for chronic pain syndrome after multiple lumbar operations. Appl Neurophysiol 45:201-204 22. Skalpe IO (1978) Adhesive arachnoiditis following lumbar myelography. Spine 3/I: 61-64 23. Smolik EA, Nash FP (1951) Lumbar spinal arachnoiditis: A complication of the intervertebral disc operation. Ann Surg 133: 490-495

151

24. Spiller W, Musser J, Martin E (1903) A case ofintradural spinal cyst with operation and recovery. University Penn Med Bull 16: 27-31 25. Thomalske G, Galow W, Ploke G (1977) Critical comments on a comparison of two series (1,000 patients each) of lumbar disc surgery. Advances in Neurosurgery 4:22-27 26. Wilkinson HA, Schuman N (1979) Results of surgical lysis of lumbar adhesive arachnoiditis. Neurosurgery 4/5:401-409

Correspondence and Reprints: Prof. Dr. med. Charles Probst, Neurochirurgische Klinik, Kantonsspital, CH-5001 Aarau, Switzerland.