|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Background Lumbar-disk surgery often is performed in patients who have sciatica that does not resolve within 6 weeks, but the optimal timing of surgery is not known.
Methods We randomly assigned 283 patients who had had severe sciatica for 6 to 12 weeks to early surgery or to prolonged conservative treatment with surgery if needed. The primary outcomes were the score on the Roland Disability Questionnaire, the score on the visual-analogue scale for leg pain, and the patient's report of perceived recovery during the first year after randomization. Repeated-measures analysis according to the intention-to-treat principle was used to estimate the outcome curves for both groups.
Results Of 141 patients assigned to undergo early surgery, 125 (89%) underwent microdiskectomy after a mean of 2.2 weeks. Of 142 patients designated for conservative treatment, 55 (39%) were treated surgically after a mean of 18.7 weeks. There was no significant overall difference in disability scores during the first year (P=0.13). Relief of leg pain was faster for patients assigned to early surgery (P<0.001). Patients assigned to early surgery also reported a faster rate of perceived recovery (hazard ratio, 1.97; 95% confidence interval, 1.72 to 2.22; P<0.001). In both groups, however, the probability of perceived recovery after 1 year of follow-up was 95%.
Conclusions The 1-year outcomes were similar for patients assigned to early surgery and those assigned to conservative treatment with eventual surgery if needed, but the rates of pain relief and of perceived recovery were faster for those assigned to early surgery. (Current Controlled Trials number, ISRCTN26872154
[controlled-trials.com]
.)
Methods
We conducted a multicenter, prospective, randomized trial among patients with 6 to 12 weeks of severe sciatica to determine whether a strategy of early surgery leads to better outcomes during the first year than does a strategy of conservative treatment for an additional 6 months followed by surgery for patients who do not have improvement. The medical ethics committees at the nine participating hospitals approved the protocol. Written informed consent was obtained from all patients. Details of the design and study protocol have been published previously.11
Eligibility and Randomization
Eligible patients were 18 to 65 years of age, had a radiologically confirmed disk herniation, and had received a diagnosis from an attending neurologist of an incapacitating lumbosacral radicular syndrome that had lasted for 6 to 12 weeks. Correlation of magnetic resonance imaging (MRI) findings with symptoms was registered by the neurosurgeon. At the time of enrollment, an independent research nurse verified the persistence of symptoms. Patients presenting with cauda equina syndrome, muscle paralysis, or insufficient strength to move against gravity were excluded. Other exclusion criteria were the occurrence of another episode of symptoms similar to those of the current episode during the previous 12 months, previous spine surgery, bony stenosis, spondylolisthesis, pregnancy, or severe coexisting disease.
A computer-generated permuted-block scheme was used for randomization, with patients stratified according to center. One hour before randomization, the patients were evaluated again, and patients who had recovered from their symptoms at that time were excluded from the trial. For patients who were included, the next numbered opaque envelope containing the assigned treatment was opened and the patient was assigned to a treatment group. The patients could not be blinded to treatment group.
Treatment
Early surgery was scheduled within 2 weeks after assignment and was canceled only if spontaneous recovery occurred before the date of surgery. The symptomatic disk herniation was removed by a minimal unilateral transflaval approach with magnification, with the patient under general or spinal anesthesia. The goal of surgery was to decompress the nerve root and reduce the risk of recurrent disk herniation by performing an annular fenestration, curettage, and removal of loose degenerated disk material from the disk space with the use of a rongeur, without attempting to perform a subtotal diskectomy. The duration of the hospital stay depended on the patient's mobility after surgery. Usual care was provided according to the protocols of the participating surgical departments. Rehabilitation of the patients at home was supervised by physiotherapists using a standardized exercise protocol. The patients were advised to resume their regular jobs when they were able, depending on the nature of their work.
General practitioners provided prolonged conservative treatment to the patients. The patients were informed about their favorable prognosis and were invited to visit the Web site for our trial, which was designed exclusively to inform patients about the natural course of their illness and the expectation of successful recovery, irrespective of the initial intensity of their pain. Treatment was aimed mainly at enabling the patients to resume daily activities. If necessary, the prescription of pain medication was adjusted according to existing clinical guidelines.11 Patients who were fearful of moving were referred to a physiotherapist. If sciatica persisted for 6 months after the patient underwent randomization, microdiskectomy was offered. Patients who had increasing leg pain not responsive to medication or progressive neurologic deficits were offered surgery earlier than 6 months after randomization.
Outcomes
The patients were assessed by means of the Roland Disability Questionnaire for Sciatica,12 the 100-mm visual-analogue scale for leg pain,13 and a 7-point Likert self-rating scale of global perceived recovery. Functional disability, intensity of leg pain, and global perceived recovery were the primary outcomes and were assessed at 2, 4, 8, 12, 26, 38, and 52 weeks.
Secondary outcomes were recorded at monitoring visits scheduled at 8, 26, and 52 weeks. At these visits the patients underwent a repeated neurologic examination; functional and economic observational assessments14 were performed by the independent research nurse; and scores on the Medical Outcomes Study 36-Item Short-Form General Health Survey (SF-36) scale,15 the Sciatica Frequency and Bothersomeness Index,12,16 and a 100-mm visual-analogue scale for health perception11 were obtained. Research nurses observed their own patients at the planned follow-up times and were not blinded to the patients' treatment assignments.
Statistical Analysis
The aims of this study were to determine the difference between the two treatment groups in disease-specific disability with respect to daily functioning, as measured by the score on the Roland Disability Questionnaire and the visual-analogue scale for leg pain, and to determine the difference in median time to recovery, measured by dichotomized self-assessment on the Likert scale as a function of time since randomization. On the assumption of a mean standard deviation of 10 points16 during the first year, we calculated that a sample of 140 patients per treatment group would be required to provide a statistical power of 0.90 with a two-tailed significance level of 0.05 to detect a difference of at least 3 points in the score on the Roland Disability Questionnaire.
Recovery was defined as complete or nearly complete disappearance of symptoms as measured on a 7-point Likert scale. Although this trial was designed primarily to determine average differences in functional outcome, it was initially estimated that this sample size would also have a statistical power of 90% to detect a difference of 2 months in the median time to recovery with the use of estimates from survival models.
Data collection and checking for quality were performed with the ProMISe data management system of the Department of Medical Statistics and BioInformatics17 of the Leiden University Medical Center. SPSS software, version 12.0,18 was used for all statistical analyses.
Differences between groups at baseline were assessed by comparing means, medians, or percentages, depending on the type of variable. The baseline values of variables were used as covariates in the main analyses, whenever appropriate, to adjust for possible differences between the randomized groups and to increase the power of the analyses. The outcomes of function and pain were analyzed with a repeated-measures analysis of variance using a first-order autoregressive covariance matrix. The estimated consecutive scores were expressed as means and 95% confidence intervals. Pointwise estimates were obtained by using models with time as a categorical covariate to allow assessment of systematic patterns.
Differences between treatment groups were assessed by estimating either the main effect of the treatment or the interaction between treatment and time. As a second approach to quantifying the differences between the two groups over total follow-up time, the areas under the curve between the time of randomization and week 52 were calculated and compared by Student's t-test. Finally, a Kaplan–Meier survival analysis was used to estimate the time elapsed from randomization until recovery, and the curves were compared with the use of a log-rank test.
A Cox model was used to compare speeds of recovery by calculation of a hazard ratio. Whether the speed of recovery differed among subgroups of patients with different characteristics11 was assessed by testing the interaction between each subgroup variable and the randomization variable, with a cutoff value of 0.10 for significance because of the lower power of the interaction test. All analyses were performed according to the intention-to-treat principle.
Results
Between November 2002 and February 2005, 599 patients meeting the criteria for surgery, as determined by their general practitioners, were contacted (Figure 1). After initial consultation with a neurologist, 395 patients who met the inclusion criteria were referred for MRI. At the second visit, 283 patients continued to have symptoms of sciatica, and disk herniation responsible for the symptoms had been observed by means of MRI. These patients were randomly assigned to one of the two treatment strategies. No significant differences were noted in baseline characteristics between patients in the two study groups (Table 1). Of 141 patients assigned to receive early surgical treatment, 16 recovered before surgery could be performed. The median time from randomization to surgery for the remaining 125 patients was 1.9 weeks (Table 1). Of the 142 patients assigned to conservative treatment, 55 underwent surgery during the first year after a median period of 14.6 weeks because of intractable pain, as measured by a mean score on the visual-analogue scale of 54 mm for leg pain and a score on the Roland Disability Questionnaire of 15.0. In the early-surgery group, 3.2% of patients had recurrent sciatica leading to a second surgical intervention, as compared with 1.8% of patients in the conservative-treatment group who underwent surgery. Complications occurred in 1.6% of all surgical patients, consisting of two dural tears and one wound hematoma. All complications resolved spontaneously. None of the patients had neurologic signs after surgery.
|
|
|
|
|
|
Although relief of symptoms was twice as fast among patients with sciatica who were treated with early surgery as among those who were treated conservatively, this multicenter, randomized trial demonstrated that this strategy did not result in a better overall 1-year functional recovery rate than did a policy of prolonged conservative treatment with an offer of subsequent surgery. During the 12 months after randomization, 89% of patients in the early-surgery group and 39% of those in the conservative-treatment group underwent microdiskectomy. At 1 year of follow-up, there were no significant differences between the groups in the mean scores for any outcome measurement, including leg pain. Thus, the major advantage of early surgical treatment is faster relief of sciatica.
The slow rate of recovery of daily functioning in the first 2 weeks after early surgery may have been due to the use of standard microdiskectomy techniques rather than modern microendoscopic or sequestrectomy methods.19,20,21 Recovery was faster during the following weeks, but there was no significant difference between the groups in the overall rate of recovery during the first year. The scores on the Roland Disability Questionnaire did not reach the minimal clinically important difference of 4 points required to conclude that early surgery results in clinically superior outcomes.11,12 Relief from leg pain occurred significantly faster in the early-surgery group, but the maximum differences between the groups in the mean scores on the visual-analogue scale for leg pain were less than 20 mm on a 100-mm scale, and at 1 year the scores were nearly equal.
The benefits of surgery for speed of recovery and relief of pain were consistent among patients in all predefined subgroups, except for patients whose sciatica was not provoked by sitting. However, the beneficial effect was marginally significant, and the majority of patients (76%) did have provocation of sciatica by sitting. It is reasonable to assume, however, that daily functioning is highly influenced by the inability to sit without pain. The absence of interactions between the assigned treatment and Lasègue's sign, intensity of pain, disk sequestrations detected by MRI, and the preference of the patient for the type of treatment was remarkable and unexpected.
Since 1934, many studies have demonstrated the success of surgical treatment of sciatica. In Weber's landmark study comparing surgery with conservative care in a randomized clinical trial, which excluded patients with "intolerable" pain, the outcome of surgery was superior at 1-year follow-up, whereas after 4 years the results of surgery and conservative treatment no longer differed.22,23,24 Surgery had some early advantages in a randomized study comparing surgery with the use of corticosteroids.25
Weinstein et al. recently reported the results of the Spine Patient Outcomes Research Trial (SPORT) comparing surgery with conservative treatment but failed to show any advantage of surgery for primary outcomes in their intention-to-treat analyses.26 Substantial crossover, however, occurred in both treatment groups, resulting in a difference in surgery rates of only 14% at 6 weeks. Furthermore, only 59% of patients assigned to surgery actually underwent surgery, which apparently was scheduled at highly variable times during the first year instead of being performed early. Another difference between our study and that by Weinstein et al. is that we enrolled patients who had had sciatica for 6 to 12 weeks, whereas at least 20% of the patients in the study by Weinstein et al. had had symptoms for at least 6 months.26
The primary outcomes of our study were also strongly influenced by a substantial crossover of patients assigned to conservative treatment, but the effects of crossover on the differences between the groups were mitigated by early performance of surgery in the group assigned to surgery. Although 61% of patients recovered quickly without surgery, the remaining 39% continued to register relatively high pain and disability scores, concordant with physical suffering for a prolonged period until surgery was performed. Österman et al. recently reported the results of a trial similar to ours showing the same trend, with earlier recovery of those assigned to surgery and nearly 40% of patients undergoing seemingly "inevitable" surgery during conservative management; however, this study did not accrue enough patients to gain adequate statistical power.27
Sciatica has high direct and indirect costs.2 Most of these costs are not generated by medical treatment but are attributed to loss of productivity. More than 1.5 million disk surgeries are performed annually worldwide, with various strategies for the timing of surgery.28 We are not aware of earlier studies that have evaluated how the timing of surgery affects outcome. Patients need a thorough understanding of the course of symptoms to inform their decisions about surgery. The results of this study will help in the decision-making process.
This study had several features that may limit the generalizability of its findings. First, patients assigned to conservative therapy were guided by research nurses who participated in pain management. Although this additional support did not prevent surgery in 39% of patients with severe sciatica, it does not reflect usual care. This must be kept in mind when a strategy of prolonged conservative treatment is implemented for wider populations. Second, it was clearly impossible for the patients and the nurses to be blinded to the treatment assignment. Finally, the time until recovery was determined on the basis of examinations performed only at predefined times during follow-up. The exact date of recovery was not determined, resulting in an underestimation of the speed of recovery in the interval between the sampling time points; however, both treatment groups were affected in the same way.
In the present study, patients with sciatica who were considering disk surgery were provided information about how early surgery and conservative treatment affect three outcome measures: disease-specific disability, intensity of leg pain, and time to recovery. Our findings suggest that patients are more likely to choose surgery if they are not able to cope with leg pain, find the natural course of recovery from sciatica unacceptably slow, and want to minimize the time to recovery from pain. Patients whose pain is controlled in a manner that is acceptable to them may decide to postpone surgery in the hope that it will not be needed, without reducing their chances for complete recovery at 12 months. Although both strategies have similar outcomes after 1 year, early surgery remains a valid treatment option for well-informed patients.
Supported by a grant from the Netherlands Organisation for Health Research and Development (ZonMW) and the Hoelen Foundation, The Hague.
No potential conflict of interest relevant to this study was reported.
We thank Mrs. Gail Bieger for reviewing the manuscript and the patients for participating in the study.
Source Information
From the Departments of Neurosurgery (W.C.P., R.T.W.M.T.), Medical Statistics (H.C.H., R.B.), Medical Decision Making (W.B.H.), and Public Health and Primary Care (J.A.H.E.), Leiden University Medical Center, Leiden; the Department of Neurology and Neurosurgery, Medical Center Haaglanden, The Hague (W.C.P., J.T.J.T.); and the Department of General Practice, Erasmus University Medical Center, Rotterdam (B.W.K.) — all in the Netherlands.
Address reprint requests to Dr. Peul at the Department of Neurosurgery, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, the Netherlands, or at w.c.peul{at}lumc.nl.
References
The participants in the Leiden–The Hague Spine Intervention Prognostic Study Group were as follows: Protocol Committee — W.C. Peul, B.W. Koes, and R.T.W.M. Thomeer; Steering Committee — B.W. Koes, R.T.W.M. Thomeer, J.A.H. Eekhof, J.T.J. Tans, W.B. van den Hout, W.C. Peul (principal investigator), R. Brand, and H.C. van Houwelingen; Statistical Analysis — R. Brand, W.C. Peul, and H.C. van Houwelingen; Manuscript Preparation — W.C. Peul, B.W. Koes, and R.T.W.M. Thomeer; Research Nurses and Data Collection and Management — M. Nuyten, P. Bergman, G. Holtkamp, S. Dukker, A. Mast, L. Smakman, C. Waanders, L. Polak, and A. Nieborg; Participating Hospitals and Coordinating Physicians — Medical Center Haaglanden, The Hague — J.T.J. Tans and R. Walchenbach; Diaconessen Hospital, Leiden — J. van Rossum, P. Schutte, and R.T.W.M. Thomeer; Groene Hart Hospital, Gouda — G.A.M. Verheul, J.E. Dalman, and J.A.L. Wurzer; Reinier de Graaf Hospital, Delft/Voorburg — J.W.A. Sven and A. Kloet; Spaarne Hospital, Heemstede/Haarlem — I.S.J. Merkies and H. van Dulken; Bronovo Hospital, The Hague — P.C.L.A. Lambrechts and J.A.L. Wurzer; Haga Hospital, The Hague — R.W.M. Keunen and C.F.E. Hoffmann; Rijnland Hospital, Leiderdorp/Alphen ad Rijn — J. Haan and H. van Dulken; Lange Land Hospital, Zoetermeer — R. Groen and R.R.F. Kuiters; Leiden University Medical Center, Leiden — R.A.C. Roos and J.H.C. Voormolen; Public Health and Primary Care, Leiden University, Leiden — J.A.H. Eekhof.
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
