Abstract
The outcome of surgery for recurrent lumbar disc herniation is debatable. Some studies show results that are comparable with those of primary discectomy, whereas others report worse outcomes. The purpose of this study was to compare the outcome of revision lumbar discectomy with that of primary discectomy in the same cohort of patients who had both the primary and the recurrent herniation at the same level and side.
A retrospective analysis of prospectively gathered data was undertaken in 30 patients who had undergone both primary and revision surgery for late recurrent lumbar disc herniation. The outcome measures used were visual analogue scales for lower limb (VAL) and back (VAB) pain and the Oswestry Disability Index (ODI).
There was a significant improvement in the mean VAL and ODI scores (both p < 0.001) after primary discectomy. Revision surgery also resulted in improvements in the mean VAL (p < 0.001), VAB (p = 0.030) and ODI scores (p < 0.001). The changes were similar in the two groups (all p > 0.05).
Revision discectomy can give results that are as good as those seen after primary surgery.
Cite this article: Bone Joint J 2013;95-B:90–4.
The incidence of sciatica in patients with lumbar disc herniation varies between studies, and depends on the definitions used.1 True radicular pain as a result of lumbar disc herniation has a favourable natural history, with surgery being reserved for unresolved or unacceptable symptoms. The timing of surgery is a matter of debate, with the only absolute indication being cauda equina syndrome. Current literature suggests that surgery in patients with primary lumbar disc herniation may be the most rapid means of resolving pain but the results at one year are similar to those achieved by non-operative measures.2-4
Lumbar discectomy remains the operative treatment for lumbar disc herniation. Although short-term outcomes are generally favourable, the long-term outcomes are more controversial.5,6 A poor outcome after lumbar discectomy occurs in between 15% and 30% of patients7-10 and is thought to be directly related to adverse prognostic factors such as an increase in back pain, persistent symptoms, or recurrence.11 Recurrent lumbar disc herniation occurs in 5% to 18% of patients12-14 and has been regarded as an objective measure of surgical failure.11,15,16 However, as the timing of recurrence varies, this may be something of a simplification. Early recurrence could be viewed as technical failure, whereas late recurrence may be constitutionally or genetically determined.
The best management of recurrent lumbar disc herniation remains controversial with debate revolving around further discectomy with or without fusion. More recently, less invasive decompressive procedures have been suggested.17,18 The low incidence of recurrent herniation, as well as its association with previous discectomy, makes randomised controlled trials difficult. Consequently, most papers that have assessed the outcome of revision lumbar discectomy compare it with that following primary discectomy.16,19-21 However, these are retrospective studies with no standardised definition of recurrence. Patients with radicular pain as a result of epidural fibrosis or lateral stenosis have also been included.16 Comparisons are made between two different cohorts of patients where the primary procedure may have been performed elsewhere.19 Many also fail to use validated spinal outcome measures.20-22 Few studies have sound methodology, which means that their results vary widely. Some report outcomes that are comparable with those of primary discectomy,10,23-25 whereas others report worse outcomes.12,26-28 The purpose of this study was to assess whether the outcome after surgery for recurrent lumbar disc herniation was favourable when compared with the primary discectomy in the same cohort of patients.
Patients and Methods
This was a retrospective analysis of prospectively gathered data about patients who, between 1996 and 2009, had undergone both a primary and revision operation for single-level lumbar disc herniation. Both procedures were performed under the care of the two senior authors (MN and PS).
Each patient had at least four to six weeks of symptoms that were unresponsive to conservative treatment, and had positive clinical and radiological (MRI) findings before both primary and revision procedures. Revision surgery was performed in patients with clinical and radiological features of late recurrent lumbar disc herniation. This was defined as the presence of clinical and radiological findings suggestive of repeat disc herniation at the same level and side as that operated on at least six months following the primary surgery. This had the practical value of having available the initial early outcome measures demonstrating an improvement after the index procedure.
We excluded patients who had only undergone either a primary or revision discectomy alone or at other institutions, those with early (< six months) recurrence, recurrence at a different level or side from the primary surgery, cauda equina syndrome, patients who failed to show improvement within five to six months after primary surgery, as well as patients whose symptoms were not related to recurrent lumbar disc herniation.
The operative technique for both procedures consisted of a mini-open posterior lumbar flavectomy, removal of the disc material and undercutting facetectomy, where required. The same approach was used in patients undergoing revision discectomy with six patients having repeat discectomy with fusion, and the remaining 24 undergoing a revision discectomy alone. Fusion was performed by an author (MN) in the earlier cases as a result of surgical preference using a polyaxial pedicle screw construct at the single affected segment. All patients were mobilised on the day after surgery and were discharged home once safe. They were subsequently reviewed according to their clinical need.
All patients completed visual analogue scores for lower limb (VAL) and back (VAB) pain and the Oswestry Disability Index (ODI)29 pre- and post-operatively for both the index and the revision procedure. After the revision surgery, patients were asked to rate their level of satisfaction according to the Macnab30 and Global31,32 satisfaction criteria.
Statistical analysis
The data were analysed using SPSS v16 (SPSS Inc., Chicago, Illinois). The change in outcome scores after both procedures was analysed using the paired samples t-test. The independent samples t-test was used to compare the outcome after primary discectomy between patients who subsequently went on to have revision surgery and those who did not have a recurrence. It was also used to compare the pre- and post-operative scores between index and revision procedures to identify whether patients were symptomatically worse pre-operatively and/or made a similar recovery post-operatively. Pearson’s correlation with linear regression analysis was used to identify any association between the individual outcome scores after primary surgery; the age at revision surgery; the interval between the two procedures; and the outcome after revision surgery.
Results
Between April 1996 and December 2009, 546 primary lumbar discectomies were carried out under the care of the two senior authors (MN and PS). Of these, 36 patients (19 men, 17 women) underwent revision surgery for recurrent disc herniation at the same level and side. This gives a revision rate of 6.6% over 13 years. Six patients (one man, five women) underwent early revision surgery for recurrent herniation (≤ six months) and were excluded. Of the 30 patients included, all had clinical improvement in outcome at five to six months after the primary discectomy. Complete VAL and ODI data were available for all 30 patients. Pre-operative VAB scores for the revision procedure were available for only 17 of 30 patients, a third of whom underwent fusion with repeat disc excision.
Recurrence occurred at the L4/5 level in 14 patients and at L5/S1 in 16. The mean age at the index discectomy was 39 years (25 to 76) and at revision surgery was 45 years (27 to 77). The mean interval between operations was 41 months (8 to 122). The mean follow-up was 16 months (5 to 80) after the index procedure and 49 months (6 to 147) after revision surgery.
Primary discectomy
Of the 546 patients who underwent a primary lumbar discectomy, complete follow-up data were available for 296 (54%). Primary lumbar discectomy resulted in a statistically significant improvement in all outcome measures assessed. These are summarised in Table I.
Table I
Mean pre- and post-operative scores following primary lumbar discectomy (CI, confidence interval)
| Mean (sd) outcome* | Mean score (sd) | p-value (t-test) |
|---|---|---|
| VAL (n = 191) | ||
| Pre-operative | 74 (19) | |
| Post-operative | 26 (26) | |
| Mean difference (95% CI) | 48 (44 to 53) | < 0.001 |
| VAB (n = 163) | ||
| Pre-operative | 48 (30) | |
| Post-operative | 26 (26) | |
| Mean difference (95% CI) | 22 (17 to 28) | < 0.001 |
| ODI (n = 296) | ||
| Pre-operative | 54 (16) | |
| Post-operative | 24 (22) | |
| Mean difference (95% CI) | 30 (28 to 33) | < 0.001 |
-
* VAL, visual analogue scale for lower limb pain; VAB, visual analogue scale for back pain; ODI, Oswestry Disability Index
Revision discectomy
For the revision procedure, the mean pre-operative scores were VAL 75 (30 to 100); VAB (50 to 90) 71 and ODI 57 (42 to 86). The mean scores at final follow-up after revision surgery were VAL 39 (0 to 100) (p < 0.001), VAB 41 (0 to 100) (p = 0.030) and ODI 31 (4 to 78) (p < 0.001). Revision discectomy resulted in a statistically significant improvement in all the outcome measures (Table II).
Table II
Mean outcome scores (with standard deviations) in the 30 patients following primary and revision discectomy (VAL, visual analogue scale for lower limb pain; VAB, visual analogue scale for back pain; ODI, Oswestry Disability Index)
| VAL (n = 30) | VAB (n = 17) | ODI (n = 30) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Pre-op | Post-op | p-value | Pre-op | Post-op | p-value | Pre-op | Post-op | p-value | |||
| Primary | 73 (16) | 41 (27) | < 0.001 | 54 (31) | 48 (35) | 0.791 | 54 (12) | 32 (25) | < 0.001 | ||
| Secondary | 75 (24) | 39 (34) | < 0.001 | 71 (19) | 41 (36) | 0.030 | 57 (12) | 31 (24) | < 0.001 | ||
| p-value | 0.746 | 0.806 | 0.083 | 0.596 | 0.473 | 0.901 | |||||
The six patients who underwent additional fusion improved similarly in all outcome measures compared with those having repeat discectomy alone, although statistical analysis could not be performed as a result of small numbers (Table III). When patients who underwent revision surgery with fusion were excluded, clinical significant improvement in back pain remained but was no longer statistically significant (p = 0.174).
Table III
Comparison of mean outcome scores following revision lumbar discectomy with or without fusion
| Outcome* | Fusion (n = 6) | Non-fusion (n = 24) |
|---|---|---|
| VAL (n = 191) | ||
| Pre-operative | 67 | 76 |
| Post-operative | 33 | 40 |
| Change | 34 | 36 |
| VAB (n = 163) | ||
| Pre-operative | 65 | 74 |
| Post-operative | 36 | 42 |
| Change | 29 | 32 |
| ODI (n = 296) | ||
| Pre-operative | 49 | 58 |
| Post-operative | 24 | 33 |
| Change | 25 | 25 |
-
* VAL, visual analogue scale for lower limb pain; VAB, visual analogue scale for back pain; ODI, Oswestry Disability Index
Outcomes for primary vs revision lumbar disc surgery (30 patients)
The mean pre-operative VAL, VAB and ODI scores for the 30 patients studied for the primary procedure were 73 (42 to 100), 54 (0 to 90) and 54 (22 to 68), respectively. The mean post-operative scores after the primary procedure were 41 (5 to 75), 48 (5 to 75) and 32 (0 to 72), respectively. After primary discectomy there was a clinical and statistical significant improvement in the VAL and ODI scores (both p < 0.001) and a small improvement in the VAB score, which was not statistically significant (p = 0.791) (Table II).
This group of patients requiring subsequent revision surgery for ‘late’ recurrent lumbar disc herniation had inferior improvements in terms of VAL pain scores following the primary discectomy when compared with the remainder of the patients not known to have had a recurrence (mean score changes of 32 and 50, respectively; p = 0.009). There were no significant differences in the pre-operative scores between these groups of patients or postoperative changes for VAB (6 vs 23, p = 0.093) and the ODI (22 vs 30, p = 0.080).
After primary discectomy there was a mean 32-point improvement in VAL score, a 6-point improvement in VAB and a 22-point improvement in ODI. By comparison, for revision discectomy there was a mean 36-point improvement in VAL (p = 0.730), a 30-point improvement in VAB (p = 0.232) and a 26-point improvement in ODI scores (p = 0.685), with no significant differences between the two procedures.
The pre- and post-operative scores for both procedures were compared to assess, first, whether patients were symptomatically worse pre-operatively after recurrent herniation, and secondly whether their outcomes were different after revision surgery compared with the index procedure. The mean results were similar, with no differences in the outcome scores for VAL and ODI (Table II; Fig. 1). The biggest differences were between the pre-operative VAB scores, with patients reporting greater levels of back pain before surgery for a recurrent herniation than before the primary procedure (71 vs 54, respectively). However, the VAB score was only available for 17 patients and failed to show any significant difference (p = 0.083).
Fig. 1
Bar charts showing the pre- and post-operative outcome scores for the 30 patients following primary and revision surgery for lumbar disc herniation, with 95% confidence intervals. VAL, visual analogue scale for lower limb pain; VAB, visual analogue scale for back pain; ODI, Oswestry Disability Index.
There was no difference between men and women in the outcome of either primary or revision discectomy, nor was there any association between the individual outcome scores after primary discectomy, the age of the patient at the time of revision or the interval between the two procedures and the outcome after revision surgery.
The complications of primary discectomy (546 patients) included 15 dural tears (2.7%); three cases of discitis; five superficial wound infections; one deep venous thrombosis; and one life-threatening polymyositis. All of these patients made a satisfactory recovery with conservative treatment; no patient with these complications developed a recurrent herniation requiring revision surgery. The complications of revision discectomy (36 patients) were four dural tears (11%) and one case of high-volume blood loss (approximately 500 ml). Again, all of these patients recovered well with conservative management.
When assessing patient-perceived global outcomes after revision discectomy, a mean 43-point change in ODI and 52-point change in VAL were associated with a ‘much better’ outcome; a 26-point ODI and 30-point VAL change with a ‘better’ outcome; a 2-point change in ODI and zero change in VAL with no change; and a minus 15-point change in ODI and a negative 22-point change in VAL with a worse outcome (Fig. 2). Overall, 74% of patients felt the outcome from revision discectomy was better or much better according to the global satisfaction criteria, with 85% of patients rating their treatment as either good or excellent according to the McNab30 satisfaction criteria.
Fig. 2
Bar charts showing the clinical important difference in scores following revision surgery for recurrent lumbar disc herniation in the study group (30 patients).
Discussion
After a primary lumbar discectomy, recurrence at the same site or an adjacent disc can occur in up to 18% of patients.12-14 The risk of recurrent herniation decreases with time to 0.6% between the second and fifth years from the primary operation, and to about 0.1% during the subsequent ten years.33 The estimated rate of reoperation for recurrent herniation has been reported to be 7.9% by ten years.21,33 The revision rate of 6.6% in our patients is therefore comparable with those of previous studies.
Recurrent lumbar disc herniation has been referred to as an ‘objective measure of surgical failure’.11,15,16 It is not clear from these studies whether this refers to early or late recurrence. All patients in this study had improved in the first five or six months after the primary discectomy and had a ‘late’ recurrence. A cut-off point of six months from surgery was felt to be ample time for any potential placebo effect to have subsided and technical failure as a cause of persistent symptoms or recurrent disc herniation to have become apparent.
Patients with a late recurrence showed significant improvements in VAL and ODI scores after their primary surgery when assessed in isolation. However, their outcome in terms of leg pain was worse than that of patients who only underwent a primary discectomy. The aim of surgery for lumbar disc herniation is to improve radicular leg pain. If judged on this basis, the results of this study suggest that late recurrence is an objective measure of surgical failure after primary discectomy.
The results of revision discectomy are satisfactory in between 50% and 90% of patients.21,34-36 Whether recurrent discectomy is as successful as primary discectomy continues to be the subject of debate. Some authors conclude that revision discectomy has a worse outcome than primary discectomy,12,26,28 whereas others conclude the opposite.10,23-25 Time is an important factor when assessing satisfaction and the outcome after surgery. Fritsch et al37 showed that the percentage of patients who had a satisfactory outcome after repeat surgery declined from 80% in the short term to 22% at long-term follow-up (> two years). Poor patient selection, perineural fibrosis and incorrect diagnosis have all been cited as possible contributory factors.
Our findings are similar to those of previous publications and show that the results of surgery for recurrent lumbar disc herniation are favourable. There were clinically significant improvements in the visual analogue scores for lower limb pain and the Oswestry Disability Index following both the primary and revision procedures. Primary discectomy failed to achieve a statistically significant improvement in back pain. Comparisons of the pre- and post-operative scores between the two procedures, despite showing noticeable differences in the pre-operative back pain scores, failed to identify any significant differences in outcome. Hence, patients who have recurrent lumbar disc herniation are no worse off pre- or post-operatively than those undergoing the primary procedure, but do report more back pain if the herniation recurs. We also failed to identify any correlation between the change in primary and revision surgery scores for all the outcomes measured. The outcomes of revision surgery are therefore independent of those for primary surgery.
This study addresses some of the major flaws present in previously published material. The patients in both groups were the same: all had clinical, radiological and surgical evidence of recurrent lumbar disc herniation on the same side and at the same level, and all had improved following primary surgery. Differences between the groups were therefore negligible, and other causes of persistent or recurrent symptoms were excluded.
The incidence of dural tears in patients undergoing revision surgery was 11% compared with 2.8% in those undergoing primary discectomy. These are comparable with previous studies: Morgan-Hough et al21 reported a rate of dural tear of 14.3% in revision and 5.5% in primary operations on the lumbar disc.
One of the limitations of this study is the relatively small number of patients. However, most studies that address the outcomes of surgery for recurrent lumbar disc herniation analyse similar numbers15,16,18-21,24,25 and larger samples could only be achieved in a multicentre trial. We also appreciate that although we have quoted a revision rate of 6.6%, some patients may have been operated on elsewhere and this may therefore be an underestimate.
1 Konstantinou K , DunnKM. Sciatica: review of epidemiological studies and prevalence estimates. Spine (Phila Pa 1976)2008;33:2464–2472. Google Scholar
2 Peul WC , van den HoutWB, BrandR, et al.Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised trial. BMJ2008;336:1355–1358. Google Scholar
3 van den Hout WB , PeulWC, KoesBW, et al.P Prolonged conservative care versus early surgery in patients with sciatica from lumbar disc herniation: cost utility analysis alongside a randomised controlled trial. BMJ2008;336:1351–1354. Google Scholar
4 Weinstein JN , TostesonTD, LurieJD, et al.Surgical vs nonoperative treatment for lumbar disk herniation: the Spine Patient Outcomes Research Trial (SPORT): a randomized trial. JAMA2006;296:2441–2450. Google Scholar
5 Loupasis GA , StamosK, KatonisPG, et al.Seven- to 20-year outcome of lumbar discectomy. Spine (Phila Pa 1976)1999;24:2313–2317. Google Scholar
6 Yorimitsu E , ChibaK, ToyamaY, HirabayashiK. Long-term outcomes of standard discectomy for lumbar disc herniation: a follow-up study of more than 10 years. Spine (Phila Pa 1976)2001;26:652–657. Google Scholar
7 Fandiño J , BotanaC, ViladrichA, Gomez-BuenoJ. Reoperataion after lumbar disc surgery: results in 130 cases. Acta Neurochir (Wien)1993;122:102–104. Google Scholar
8 Mobbs RJ , NewcombeRL, ChandranKN. Lumbar discectomy and the diabetic patient: incidence and outcome. J Clin Neurosci2001;8:10–13. Google Scholar
9 Suk KS , LeeHM, MoonSH, KimNH. Recurrent lumbar disc herniation: results of operative management. Spine (Phila Pa 1976)2001;26:672–676. Google Scholar
10 Swartz KR , TrostGR. Recurrent lumbar disc herniation. Neurosurg Focus2003;15:E10. Google Scholar
11 Kim KT , ParkSW, KimYB. Disc height and segmental motion as risk factors for recurrent lumbar disc herniation. Spine (Phila Pa 1976)2009;34:2674–2678. Google Scholar
12 Ebeling U , KalbarcykH, ReulenHJ. Microsurgical reoperation following lumbar disc surgery: timing, surgical findings, and outcome in 92 patients. J Neurosurg1989;70:397–404. Google Scholar
13 Crock HV . Observations on the management of failed spinal operations. J Bone Joint Surg [Br]1976;58-B:193–199. Google Scholar
14 Law JD , LehmanRA, KirschWM. Reoperation after lumbar intervertebral disc surgery. J Neurosurg1978;48:259–263. Google Scholar
15 Guo JJ , YangH, TangT. Long-term outcomes of the revision open lumbar discectomy by fenestration: a follow-up study of more than 10 years. Int Orthop2009;33:1341–1345. Google Scholar
16 Cinotti G , RoysamGS, EisensteinSM, PostacchiniF. Ipsilateral recurrent lumbar disc herniation: a prospective, controlled study. J Bone Joint Surg [Br]1998;80-B:825–832. Google Scholar
17 Baba H , ChenQ, KamitankK, ImuraS, TomitaK. Revision surgery for lumbar disc herniation: an analysis of 45 patients. Int Orthop1995;19:98–102. Google Scholar
18 Fu TS , LaiPL, TsaiTT, et al.Long term results of disc excision for recurrent lumbar disc herniation with or without posterolateral fusion. Spine (Phila Pa 1976)2005;30:2830–2834. Google Scholar
19 Papadopoulos EC , GirardiFP, SandhuHS, et al.Outcome of revision discectomies following recurrent lumbar disc herniation. Spine (Phila Pa 1976)2006;31:1473–1476. Google Scholar
20 Acharya KN , NathanTS, KumarJR, MenonKV. Primary and revision lumbar discectomy: a three-year review from one center. Ind J Orthop2008;42;178–181. Google Scholar
21 Morgan-Hough CV , JonesPW, EisensteinSM. Primary and revision lumbar discectomy: a 16 year review from one centre. J Bone Joint Surg [Br]2003;85-B:871–874. Google Scholar
22 Palma L , CarangeloB, MuziiVF, et al.Microsurgery for recurrent lumbar disc herniation at the same level and side: do patients fare worse?: experience with 95 consecutive cases. Surg Neurol2008;70:619–621. Google Scholar
23 Jönsson B , StrömqvistB. Repeat decompression of lumbar nerve roots: a prospective two-year evaluation. J Bone Joint Surg [Br]1993;75-B:894–897. Google Scholar
24 Dai LY , ZhouQ, YaoWF, ShenL. Recurrent lumbar disc herniation after discectomy: outcome of repeat discectomy. Surg Neurol2005;64:226–231. Google Scholar
25 Suk KS , LeeHM, MoonSH, KimNH. Recurrent lumbar disc herniation: results of operative management. Spine (Phila Pa 1976)2001;26:672–676. Google Scholar
26 O’Sullivan MG , ConnollyAE, BuckleyTF. Recurrent lumbar disc protrusion. Br J Neurosurg1990;4:319–325. Google Scholar
27 Skaf G , BouclaousC, AlarajA, ChamounR. Clinical outcome of surgical treatment of failed back surgery syndrome. Surgical Neurol2005;64:483–489. Google Scholar
28 Silvers HR , LewisPJ, AschHL, ClabeauxDE. Lumbar diskectomy for recurrent disk herniation. J Spinal Disord1994;7;408–419. Google Scholar
29 Fairbank JC , PynsentPB. The Oswestry Disability Index. Spine (Phila Pa 1976)2000;25:2940–2952. Google Scholar
30 Macnab I . Negative disc exploration: an analysis of the causes of nerve-root involvement in sixty-eight patients. J Bone Joint Surg [Am]1971;53-A:891–903. Google Scholar
31 Barber BL , SantanelloNC, EpsteinRS. Impact of the global on patient perceivable change in an asthma specific QOL questionnaire (1996). Qual Life Res1996;5:117–122. Google Scholar
32 Beurskens AJ , de VetHC, KökeAJ, van der HeijdenGJ, KnipschildPG. Measuring the functional status of patients with low back pain: assessment of the quality of four disease-specific questionnaires. Spine (Phila Pa 1976)1995;20:1017–1028. Google Scholar
33 Gaston P , MarshallRW. Survival analysis is a better estimate of recurrent disc herniation. J Bone Joint Surg [Br]2003;85-B:535–537. Google Scholar
34 Bernard TN Jr. Repeat lumbar spine surgery: factors influencing outcome Spine (Phila Pa 1976)1993;18:2196–2200. Google Scholar
35 Cinotti G , GuminaS, GiannicolaG, PostacchiniF. Contralateral recurrent lumbar disc herniation: results of discectomy compared with those in primary herniation. Spine (Phila Pa 1976)1999;24:800–806. Google Scholar
36 Connolly ES . Surgery for recurrent lumbar disc herniation. Clin Neurosurg1992;39:211–216. Google Scholar
37 Fritsch EW , HeiselJ, RuppS. The failed back surgery syndrome: reasons, intraoperative findings, and long term results: a report of 182 operative treatments. Spine (Phila Pa 1976)1996;21:626–633. Google Scholar
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
