Study Design. A cohort of 100 patients with symptomatic lumbar spinal stenosis, characterized in a previous article, were given surgical or conservative treatment and followed for 10 years.
Objectives. To identify the short- and long-term results after surgical and conservative treatment, and to determine whether clinical or radiologic predictors for the treatment result can be defined.
Summary of Background Data. Surgical decompression has been considered the rational treatment. However, clinical experience indicates that many patients do well with conservative treatment.
Methods. In this study, 19 patients with severe symptoms were selected for surgical treatment and 50 patients with moderate symptoms for conservative treatment, whereas 31 patients were randomized between the conservative (n = 18) and surgical (n = 13) treatment groups. Pain was decisive for the choice of treatment group. All patients were observed for 10 years by clinical evaluation and questionnaires. The results, evaluated by patient and physician, were rated as excellent, fair, unchanged, or worse.
Results. After a period of 3 months, relief of pain had occurred in most patients. Some had relief earlier, whereas for others it took 1 year. After a period of 4 years, excellent or fair results were found in half of the patients selected for conservative treatment, and in four fifths of the patients selected for surgery. Patients with an unsatisfactory result from conservative treatment were offered delayed surgery after 3 to 27 months (median, 3.5 months). The treatment result of delayed surgery was essentially similar to that of the initial group. The treatment result for the patients randomized for surgical treatment was considerably better than for the patients randomized for conservative treatment. Clinically significant deterioration of symptoms during the final 6 years of the follow-up period was not observed. Patients with multilevel afflictions, surgically treated or not, did not have a poorer outcome than those with single-level afflictions. Clinical or radiologic predictors for the final outcomewere not found. There were no dropouts, except for 14 deaths.
Conclusions. The outcome was most favorable for surgical treatment. However, an initial conservative approach seems advisable for many patients because those with an unsatisfactory result can be treated surgically later with a good outcome.
The clinical entity “lumbar spinal stenosis” can be defined as a narrowing of the lumbar spinal canal, resulting in symptoms and signs caused by entrapment and compression of intraspinal vascular and nervous structures. Herniated intervertebral discs and space-occupying lesions (caused by inflammation or neoplasm) are in the strictest sense also causes of stenosis, but they usually are regarded as separate entities. Clinical and radiologic features are described and discussed in a previous article. 4 Pain in the back and leg(s) and, in particular, claudication caused by compression and ischemia of nerve roots are the main symptoms.
Lumbar spinal stenosis has been known for more than 100 years, but for a long time it was regarded as “the forgotten spinal disease.” This neglect occurred because the association between herniated vertebral discs and sciatica received most of the attention after it was discovered by Mixter and Barr 37 in 1934. However, since the early 1950s, starting with the studies of Verbiest, 52 this has changed, and lumbar spinal stenosis now is an accepted clinical entity. The space in the vertebral canal is limited, usually because of degenerative changes, supposed to be progressive, and sometimes in combination with a congenital narrow bony canal. Symptoms and signs are related to this limited canal space. Surgical decompression was considered as the natural treatment, and the results of this were reported in several publications. 7,9,13,14,23,25,38,52,53,57 However, over time, clinical experience indicated that many patients obviously did well without surgery. 10,19,24,35,41,42,44
Medical publications deal largely with results after surgical treatment. In a meta-analysis by Turner et al, 51 successful results after surgical treatment are reported for 26% to 100% of the subjects, but the mean follow-up time was less than 4 years. Herno et al 18 reported good surgical results for 68% of patients after a mean follow-up time of 12 years. Postacchini et al 43 examined the surgical long-term results, making a distinction between results from patients with short-term unsatisfactory and satisfactory outcomes. After an average follow-up time of 8 years the proportion of unsatisfactory results was 33%, but this decreased to 20% when only the patients who had satisfactory short-term results were considered. None of the patients with an unsatisfactory short-term result improved with time. Literature dealing with results after nonsurgical treatment are scanty.
In a study comparing surgically and conservatively treated patients, Johnsson et al 25 found that 60% of the patients treated surgically improved and 25% deteriorated, whereas of the conservatively treated patients, 30% improved and 60% were unchanged. Mean observation time was 53 and 31 months, respectively. In another study Johnsson et al 24 studied the natural course of spinal stenosis and found after a mean observation period of 49 months that the symptoms were unchanged in 70%, improved in 15%, and worsened in 15% of the patients. No proof of deterioration was found after 4 years, and it was concluded that expectant observation could be an alternative to surgical treatment.
Controlled clinical studies comparing conservative and surgical treatment are rare, and few studies deal with long-term results. 23,51 In an effort to shed light on this important problem, the current study was initiated in 1983 under the auspices of Dr. Henrik Weber and in close collaboration with the departments of neuroradiology and neurosurgery at Ullevål Hospital, Oslo, Norway. It has been influenced heavily by the experience Dr. Weber attained with patients who have lumbar disc herniation, as presented in his doctoral thesis in 1978. 55,56
The first task was to define a study population and to characterize the clinical features and radiologic aspects of patients with symptomatic lumbar spinal stenosis. For this purpose, 100 patients were recruited over a 3-year period from December 1984 to September 1987. The patients fulfilled certain criteria that reflected the prevailing definition of lumbar spinal stenosis. The results from this initial part of the research project are presented in a previous article. 4
The core of the problem, addressed in this report, is to find guidelines applicable for the individual patient on how to choose between conservative and surgical treatment. The working hypothesis of the authors was that conservative treatment is a realistic alternative to surgery, at least under certain circumstances. They defined the following aspects of the problem:
Conservative treatment seems to be the natural choice when symptoms are mild. Little is known about the short- and long-term prognosis of this approach. How often does conservative treatment give an acceptable result? Will the clinical condition deteriorate as the patient ages? This might be anticipated if a simple causal relation exists between symptoms and degenerative changes of the spine.
When symptoms are severe, it is tempting to suggest surgery. Decompression seems logical and has the potential to give the patient immediate relief. How effective is surgical treatment? What is the price for a beneficial long-term effect of surgery? Will spinal “instability” after surgery take its toll over the years in the form of an acceleration of increasing symptoms? Which symptoms benefit most from surgery?
An important problem pertains to the timing of surgery. Is an early operation important to secure a good result, or should all patients initially be offered conservative treatment, with surgery reserved for the failures of such treatment?
Is it possible to find clinical or radiologic predictors that can be used as guidelines in the decision between conservative and surgical treatment?
From the *Department of Neurology, †Department of Neurosurgery, ‡Clinical Research Forum, and §Department of Neuroradiology, Ullevål Hospital, Oslo, Norway.
Supported by the Clinical Research Forum, Ullevål University Hospital, Oslo, Norway.
Acknowledgment date: June 11, 1998.
First revision date: November 25, 1998.
Second revision date: May 17, 1999.
Acceptance date: July 17, 1999.
Address reprint requests to
Tom Amundsen, MD
0407 Oslo, Norway
Device status category: 1.
Conflict of interest category: 12, 14.