The goal of the surgeon managing lumbar spinal stenosis (LSS) should be to achieve adequate decompression of neural structures while minimizing surgical damage to posterior stabilizing structures. Traditional bilateral laminectomy is burdened by postoperative complications, such as vertebral instability, severe atrophy of paraspinal muscles, and posterior scarring, which can lead to poor results (1–3). The choice to perform multilevel total laminectomy may seem somewhat disproportionate in patients affected by mild to moderate leg pain but no severe nervous disturbances. In such cases, the most appropriate treatment, surgical or conservative, may not be obvious. To address this problem, less invasive surgical approaches should be adopted. Unilateral laminectomy was first presented as a bilateral decompressive method in cervical spondylotic myelopathy (4). This approach was recently proposed again for the management of LSS (5–7). Unilateral laminectomy preserves the facet joints and neural arch of the contralateral side, limits postoperative destabilization, and protects the nervous structures against posterior scarring. In the current prospective study, we compare single- and multiple-level unilateral laminectomy with conservative treatment in patients affected by LSS who have mild to moderate leg pain.
From 1993 to 1998, we enrolled 44 patients who were hospitalized for degenerative or combined LSS. They fulfilled the following inclusion criteria: 1) leg pain of mild to moderate intensity with unilateral radiation, for which surgical or conservative treatment could represent fair indications; 2) age more than 40 years; and 3) narrowing of the central portion of the spinal canal with an area of the dural sac measuring less than 130 mm2 at one or more intervertebral levels. The exclusion criteria were: 1) pain too severe to be treated conservatively yet too light to be treated surgically; 2) age less than 40 years; 3) congenital stenosis; 4) pure lateral stenosis with a dural sac area at the most stenotic level larger than 130 mm2; and 5) previous back surgery. Based on these criteria, intensity and unilateral distribution of the radiating pain were the critical parameters for inclusion in this study. Some patients enrolled in the current study were also included in a previous report (8). The participants (30 women, 14 men), whose mean age was 61 years (range, 41 to 75 years), were assigned to two groups: surgical (n = 22) or conservative (n = 22) treatment. Thirty-two of 44 (73%) patients were randomly assigned to the two groups by block randomization using computer-generated random numbers. In the remaining 12 cases, the participants refused randomization and chose their own treatment group (8 conservative, 4 surgical). After informed consent was obtained, all patients underwent a comprehensive interview concerning their personal and medical histories. Subsequently, complete physical and neurologic examinations were performed, and a resident unaware of the treatment assignments attributed a score to each patient for each variable of the Beaujon Scoring System (BSS) (9). The BSS is a seven-item questionnaire specifically designed for the functional assessment of patients with LSS (Table 1).
The patients were asked to evaluate and report their symptoms, need for medications, and possible limitation of daily activities of living, choosing the one statement in each item of the questionnaire that described their current limitations most accurately. Each statement described a lesser degree of impairment than the preceding statement. By adding the scores of single items, each patient's functional status can be evaluated comprehensively. The maximum value was 20, indicating the best functional status. A complete roentgenographic study, including standard and standing functional views in maximum flexion and extension of the lumbar spine, was made subsequently for each patient. The occurrence and degree of degenerative spondylolisthesis and possible instability were evaluated according to White and Panjabi (10). All patients underwent computed tomography, magnetic resonance imaging, or both. The dural sac cross-sectional areas at the level of the most stenosis were calculated using a method described in a previous report (8). Following the precedent set by other authors (11,12), we judged a level to be stenotic if the dural area measured less than 130 mm2.
Clinically defined symptoms and preoperative diagnostic imaging studies determined the choice of decompression level in the 22 patients we selected for surgical treatment. One surgeon performed all the operations. Single- or multiple-level unilateral laminectomies with complete removal of ligamenta flava on the symptomatic side were performed, and adequate lateral decompression was achieved by ipsilateral partial facetectomy to completely relieve all mechanical compression of the nerve roots. Right-sided decompression was performed in 10 patients and left-sided decompression in 12. No vertebral fusion was performed in any case, and discectomy was performed in patients with herniated disks. Bulging disks were left in place. We undercut the contralateral laminae only in cases of the most severe narrowing of the spinal canal (i.e., with dural sac cross-sectional areas measuring less than 75 mm2). To prevent postoperative instability, we performed sparing facetectomy but never discectomy in patients with degenerative spondylolisthesis. Two patients (9%) had one decompressed level, 13 (59%) patients had two, 6 (27%) patients had three, and 1 (5%) patient had four decompression levels. All were allowed to walk on the first postoperative day wearing a semirigid orthosis. One month later, use of the orthosis was discontinued. After baseline assessment, the 22 patients selected for conservative treatment were discharged with a prescription of bed rest, discontinuous use of a lumbar semirigid orthosis, physiotherapy, and an appropriate exercise program.
The patients were assessed clinically and roentgenographically after 1 year, 2 years, and at the last follow-up visit, which was performed an average of 47 months (range, 30 to 96 months) after the baseline evaluation. During each follow-up physical and neurologic examination, we made a functional assessment using the BSS and a roentgenographic study that included standard and standing functional views of the lumbar spine. During the 1-year and latest follow-up visits, we recorded the patient's degree of satisfaction regarding his pain, functional status, and treatment received. We used each patient's reported degree of satisfaction regarding pain, function, and treatment, along with the change in the BSS as the parameters to separate the results into one of the following categories: excellent (fully satisfied patient, complete or nearly complete resolution of pain and physical function impairment); fair (satisfied patient, partial resolution of symptoms); or unchanged or worse (dissatisfied patient, no change or worsening of symptoms). We considered a change in the BSS equal to or more than two points notable. Excellent and fair were considered good results, whereas unchanged or worse represented poor results. Crossovers from conservative treatment to surgery and repeated operations were also considered poor results. We obtained postoperative computed tomographic or magnetic resonance images from 17 of the 22 surgical patients. From these images, we calculated the postoperative dural sac cross-sectional area at the level of the most stenosis to obtain the grade of residual stenosis.
We used two-sample Student t tests, analyses of variance, and nonparametric tests to determine cross-sectional differences between therapeutic groups at baseline and during follow-up assessments. For each group, we used general linear models for repeated measures to evaluate the significance of longitudinal variations in global functional status, using the BSS. We also used the same method for single items on the BSS to evaluate the outcome of neurogenic claudication, leg pain at exertion, leg pain at rest, low back pain, neurologic deficit, use of pharmaceutical agents, and activities of daily living after surgical or conservative treatment.
The BSS was treated as a discrete variable, whereas the other outcomes were fitted into four categories, except for leg pain at exertion and neurologic deficits, which were assessed with three-step scales.
For all outcome variables, the higher the score, the better the result. We cross-tabulated the results at the 1-year and latest follow-up with the treatment received using dichotomous variables (good or bad), and we performed a χ2 test to determine the significance of differences between the two groups. This was the only case in which we used a χ2 test to analyze changes in BSS. Furthermore, postoperative variations in the degree of degenerative spondylolisthesis and an increase in the dural sac cross-sectional area were analyzed in the surgically treated patients. We performed univariate linear regression and multiple linear regression analyses to identify those baseline variables that were correlated significantly with the final outcome of global functional status according to BSS. We expressed variations in this parameter as a continuous variable and calculated them by subtracting the preoperative BSS from its value on the latest follow-up. The baseline explanatory variables that we included in the analysis were age, sex, body mass index, comorbid conditions affecting walking ability (categorical variable: 0 = none, 1 = 1, ≥2 = 2), number of stenotic levels, symptom duration, degree of degenerative spondylolisthesis (if any), and dural sac cross-sectional area at the level of the most stenosis. To improve the statistical power of the tests, we considered surgically and conservatively treated patients jointly, adjusting possible relations by way of treatment. When we noted a trend toward an association with the outcome of interest at the univariate analysis, we checked the explanatory variables with multiple linear regression analysis and then excluded them from analysis if they were not correlated significantly with the outcome and their inclusion did not result in notable variations of coefficients of other variables that would suggest that they played a role as confounders. Probability values less than 0.05 were considered significant. We used EPI-INFO version 5 (World Health Organization, Geneva, Switzerland) and SPSS (SPSS, Inc., Chicago, IL, U.S.A.) version 8 software programs for data collection and statistical analyses.
Baseline Characteristics of the Therapeutic Groups
Table 2 summarizes the baseline characteristics of patients in the surgical and nonsurgical groups. We noted no significant differences between the two groups for age, sex, body mass index, comorbid conditions, working activity, clinical and functional status according to the BSS, or the frequency of a positive result of a straight-leg-raising test. The only baseline parameter that differentiated the two groups was neurologic impairment, which was significantly more severe in the surgical group (p < 0.01). With respect to the aspects of disease revealed by the imaging tests, number of levels of stenosis, severity of stenosis assessed through the calculation of dural-sac cross-sectional area at the level of the most stenosis, and occurrence and degree of degenerative spondylolisthesis were not significantly different for the two groups. The most frequently affected levels were L3–L4 in 41 of 44 (95%) patients, L4–L5 in 39 of 44 (89%), L2–L3 in 16 of 44 (36%), and L5–S1 in 9 of 44 (20%). We detected no vertebral instability in any patient.
All patients were assessed during a 1-year follow-up visit. Three patients left the study during the second year: one woman in the surgical group and one man and one woman in the conservative group. A disk became herniated in the surgical patient at a decompressed level and she underwent another operation. One woman who was treated conservatively died of unrelated causes 18 months after the baseline evaluation. One man in the conservative group was reconsidered for surgery 13 months after his baseline evaluation because of disabling pain. This case was considered a poor result of conservative treatment in our analyses. Therefore, because no other patients dropped out, 41 patients were available for the latest follow-up, which was held (on average) 4 years after the baseline evaluation. Table 3 shows the clinical evolution of the patients, with signs and symptoms rated according to the BSS. The mean functional status at 1 year was significantly improved, regardless of the treatment performed. At the 2-year follow-up visits, stable improvement of the clinical scores was noted; there was a slight nonsignificant decrease in the BSS of the conservative group compared with the 1-year follow-up scores. At the time of the latest follow-up, global functional status had deteriorated compared with the 2-year result in both groups of patients. However, the surgical group had significantly sustained improvement compared with baseline levels, whereas the average BSS of the conservative group had decreased to the baseline value. On the whole, at the 2-year evaluation, and especially at the last follow-up visits, the patients treated with surgery did better than the conservatively treated patients, as measured by global and single clinical outcomes. Furthermore, neurologic deficits in these patients showed a significant and stable amelioration with respect to their baseline status, whereas we noted significant worsening after the 2-year follow-up among the conservatively treated patients. The postoperative improvement in activities of daily living was sustained over time. With regard to the results fitted in categorical variables, on the 1-year follow-up we noted good results in 82% of the surgically treated group and 68% of the conservatively treated group and bad results in 18% and 32%, respectively. After 4 years, the good results were 68% and 33%, and the bad results were 32% and 67%, respectively. This difference was significant (p < 0.01).
On the last control visit, 76% of the surgically treated patients and 30% of the conservatively treated patients were satisfied with their treatment and results. In addition to the above-mentioned cases, two more patients, one from each group, were readmitted for an operation on the latest follow-up, leading to a global repeated operation rate of 9% (2 of 22) and a crossover rate from conservative to surgical treatment of 9% (2 of 22). Residual stenosis, which was treated through contralateral decompression, was responsible for the repeated operation, whereas disabling pain prompted the crossover to surgery. The single surgical complication during the study was a dural tear that was sutured during operation with no clinical consequences.
Surgical decompression yielded significant widening of the spinal canal. In the patients who underwent postoperative imaging, the preoperative dural sac cross-sectional area at the level of the most stenosis was 70.76 ± 28.2 mm2, whereas on postoperative scans it was 108.12 ± 31.5 mm2, with an average correction rate of 65% (Fig. 1; p < 0.001). The most stenotic postoperative area at any level was more than 130 mm2 in 5 patients and more than 100 mm2 in 13 patients (of 17). Thus, although a significant enlargement of the area was obtained at all decompressed levels, we frequently observed some degree of residual or adjacent stenosis. A nonsignificant progression of degenerative spondylolisthesis over time was observed in both surgically (Fig. 2) and conservatively treated patients, with a mean increase of 2.6% and 1.9%, respectively (p > 0.05). On the latest follow-up examination, we detected neither new cases of degenerative spondylolisthesis nor any evidence of instability in any patient, regardless of the method of treatment.
Our univariate regression analysis of the BSS responses indicated that the degree of comorbidity influencing walking ability was the only significant negative outcome predictor (p < 0.05). A negative outcome was not predicted by age, sex, heavy workload, mean symptom duration, body mass index, number of stenotic levels, most stenotic dural sac cross-sectional area at baseline, or the presence of degenerative spondylolisthesis. When we performed multivariate regression analysis, the degree of comorbidity was still a significant outcome predictor (coefficient = -0.284; p < 0.05), and female patients tended to have poorer functional results (regardless of the method of treatment), although this was not significant.
Even when sound decompression has been achieved, postoperative pain and instability may worsen the results of traditional bilateral laminectomy. Postlaminectomy scar tissue formation and paraspinal muscle atrophy may cause pain in failed back syndrome, according to Sihvonen et al. (3). Furthermore, because midline structures (i.e., supraspinous and interspinous ligaments) play a mechanical role in the movement of the lumbar spine (13,14), their ablation may contribute to the pathogenesis of postoperative instability, particularly when bilateral facetectomy has been performed simultaneously. Roentgenographic instability can occur after operation without detectable preoperative instability, or in the presence of dynamic hypermobility of the stenotic segments on preoperative radiographs; such instability may lead to deterioration of surgical results and the need for repeated operation (1,15). Even a minor degree of instability accompanied by residual stenosis or scar tissue formation with adhesion to nerve roots and dura may be sufficient to cause the symptoms (2,16). Therefore, surgical techniques other than laminectomy have been introduced to reduce injury to posterior elements. Several authors have proposed using a unilateral approach to obtain decompression of neural structures in LSS (5–7).
Advantages include conservation of midline structures and contralateral lamina, facet joint, and paraspinal muscles. This results in a reduction of postoperative instability and formation of posterior scar tissue that could impinge on neural structures. Because it represents a minimally invasive procedure, it may be appropriate for patients with moderate radiologic signs or clinical symptoms of LSS, for whom conservative treatment could be an alternative. Indeed, most of these patients have stable outcomes over time after conservative treatment (17–21), and to perform traditional bilateral laminectomy in this setting may be excessive intervention. We began the current longitudinal study to compare the results of unilateral laminectomy with those of conservative treatment in cases of degenerative LSS for which either surgical or nonsurgical treatment might be appropriate. We compared patients with similar characteristics at baseline, matching them on parameters that several studies have reported could affect the outcome (1,8,9,21–29). The only parameter that significantly differentiated the two groups at enrollment was the degree of neurologic impairment: The patients undergoing surgical treatment showed more severe deficits at baseline than did the conservatively treated patients. However, because superior neurologic outcomes were observed for the surgically treated patients, this difference can be dismissed as a selection bias. Global functional status as assessed by the BSS at the 4-year follow-up visit was significantly better in the surgically treated than in the conservatively treated patients, whose mean score was similar to their baseline value. At this interval, the functional score in surgically treated patients was higher than the baseline value, even though a decline became evident subsequent to the 2-year follow-up visit. We observed similar trends for the single variables we measured with the BSS.
Our results correspond with those of some studies (21) but differ from those of others that report similar outcomes for surgically and conservatively treated patients (17,19). However, these latter studies failed to match the patients for baseline pain severity, disability, or radiologic findings, and therefore their conclusions cannot be generalized. After 4 years, we noted good results in 68% of the surgically treated and 35% of the conservatively treated patients and bad results in 32% and 65%, respectively. Our results are consistent with previous studies, some lasting as long as 10 years, that have reported good or excellent results in approximately 50% to 75% of patients treated with different methods of surgical decompression (1,5,7,15,16,20,26,28,30–39). Some of these investigators also noted a decrease in the rate of good surgical outcomes on later follow-up visits (20–23,26,31,33). Johnsson et al. (40) noted deterioration of the neurologic status for conservatively treated patients; they reported progression of neurophysiologic changes in 38% of their patients, but they considered the effect of surgery on neurologic damage controversial. Several studies (15,27) have reported improvement of neurologic deficits after surgery, whereas one did not (17). Different degrees of baseline neurologic deficit and variable diagnostic methods might explain this disagreement. Indeed, in LSS the rate of postoperative recovery varies according to the pathogenesis and duration of neurologic damage, because the short compression caused by a coexisting discal herniation in younger patients has a better prognosis than does the long-term bony compression found in elderly patients (8,27).
Our repeated operation rate of 9% was close to the average of the values reported in the literature (15,21,24,26,32,37,41,42). Patients with stenosis may undergo a second operation for different reasons, including development of disk herniation at the level of decompression (or adjacent to it), instability, or residual or recurrent stenosis (15,24,26,37,41). We repeated two operations: a discectomy for disk herniation and a bilateral laminectomy in a patient with contralateral symptoms. In both cases, the surgical technique at the time of the primary procedure could have been the cause. The presence of a markedly bulging disk without narrowing of the intervertebral space should have been treated by additional discectomy, because unilateral laminectomy in this setting may slightly increase segmental motion, possibly leading to disk herniation (15). The second repeated operation could have been avoided if preoperative contralateral leg pain and severe bilateral changes on imaging tests had been given sufficient weight. We believe that unilateral laminectomy is not suitable for cases of LSS with obvious bilateral symptoms.
It is noteworthy that we detected neither significant progression nor any new cases of degenerative spondylolisthesis after surgery in our patients. Furthermore, we found no postoperative segmental instability on functional radiographs. Degenerative spondylolisthesis, the progression of which is seen frequently on postoperative radiographic films after traditional laminectomy, has not been associated with poor results (9,15,32,43,44). However, postoperative segmental hypermobility or destabilization, either after traditional laminectomy or multiple bilateral laminotomies, can result in poorer outcomes and repeated operation (1,15,23,31,43). Unilateral laminectomy, which limits the destruction of posterior elements, also prevents postoperative segmental instability and may eliminate the need for an associated fusion. In those cases in which postoperative images were obtained, we noted a significant enlargement of the spinal canal, with a 65% increase in the dural sac cross-sectional area at the level of the most stenosis. To the best of our knowledge, ours is the first study to show an enlargement of the spinal canal and an increase in dural sac cross-sectional area after unilateral laminectomy in patients with LSS. Consistent with other studies that have reported rates of positive findings of postoperative computed tomographic or magnetic resonance imaging in 62% to 73% of surgically treated patients (16,45–47), we frequently noted slight to moderate residual or adjacent stenosis on postoperative scans. However, this finding did not correspond with poorer results. This lack of correlation might be explained by the fact that the different levels of stenosis generate different patients' symptoms (16,47). Thus, focal decompression of the critical zones may be sufficient to cause relief of symptoms, even in the presence of stenosis on postoperative scans.
As for outcome predictors, the degree of comorbidity influencing patients' ability to walk was the only baseline parameter that negatively affected the results in both surgically and conservatively treated patients. This result corresponds with those reported in other studies (23–26,29,39).
The primary limitations of our study are lack of randomization between surgically and conservatively treated patients in 27% of the cases and the small sample sizes. However, the lack of randomization did not result in different baseline characteristics of the final therapeutic groups, except for neurologic deficits. Therefore, we cannot exclude the possibility that the different outcomes observed in the two groups could be influenced by baseline differences in the neurologic status. However, if this is the case, we should conclude that, because the surgically treated patients had more severe baseline neurologic involvement and more favorable functional outcomes, we should expect even better results in the surgically treated patients in the event of fully comparable baseline characteristics of the two groups. We sacrificed sample size for more stringent inclusion criteria. However, the small sample size did not obscure longitudinal differences observed between the two therapeutic groups. Conversely, the strengths of our study are its prospective design, thorough baseline matching of the therapeutic groups, thorough imaging assessment, and long observational length.
In conclusion, two results of our study are especially noteworthy. Unilateral laminectomy in LSS leads to better 4-year results than does conservative treatment of patients who have mild to moderate leg pain with an exclusively or primarily unilateral distribution. This surgical procedure results in substantial widening of the spinal canal, as demonstrated by postoperative increases in the cross-sectional area of the dural sac. These goals can be reached without the risks for postoperative segmental instability or formation of posterior scar tissue, which may cause preoperative symptoms to recur. Because unilateral laminectomy is a minimally invasive technique, it may be appropriate for patients with moderate symptoms of LSS.
1. Johnsson KE, Redlund-Johnell I, Uden A, et al. Preoperative and postoperative instability in lumbar spinal stenosis
. Spine 1989; 14:591–3.
2. Sano S, Yokokura S, Nagata Y, et al. Unstable lumbar spine without hypermobility in postlaminectomy cases. Mechanisms of symptoms and effect of spinal fusion with and without spinal instrumentation. Spine 1990; 15:1190–7.
3. Sihvonen T, Herno A, Paljarvi L, et al. Local denervation atrophy of paraspinal muscles in postoperative failed back syndrome. Spine 1993; 18:575–81.
4. Gui L, Davidovits P, Merlini L, et al. Cervical myelopathy. Technique and results of spinal cord decompression by lateral laminectomy
. Ital J Orthop Traumatol 1983; 9:129—38.
5. diPierro CG, Helm GA, Shaffrey CI, et al. Treatment of lumbar spinal stenosis
by extensive unilateral decompression and contralateral autologous bone fusion: Operative technique and results. J Neurosurg 1996; 84:166–73.
6. Spetzger U, Bertalanffy H, Reinges MH, et al. Unilateral laminotomy for bilateral decompression of lumbar spinal stenosis
. Part II: Clinical experiences. Acta Neurochir 1997; 139:397–403.
7. Mackay DC, Wheelwright EF. Unilateral fenestration in the treatment of lumbar spinal stenosis
. Br J Neurosurg 1998; 12:556–8.
8. Mariconda M, Zanforlino G, Celestino GA, et al. Factors influencing the outcome
of degenerative lumbar spinal stenosis
. J Spinal Disord 2000; 13:131–7.
9. Lassale B, Bitan F, Bex M, et al. Résultats fonctionnels et facteurs de pronostic du traitement chirurgical des sténoses lombaires dégénératives. Rev Chir Orthop 1988; 74:85–8.
10. White AA III, Panjabi MM. Clinical Biomechanics of the Spine
. Philadelphia: Lippincott, 1990.
11. Bolender NF, Schönström NSR, Spengler DM. Role of computed tomography and myelography in the diagnosis of central spinal stenosis. J Bone Joint Surg [Am] 1985; 67:240–6.
12. Willén J, Danielson B, Gaulitz A, et al. Dynamic effects on the lumbar spinal canal. Axially loaded CT-myelography and MRI in patients with sciatica and/or neurogenic claudication. Spine 1997; 22:2968–76.
13. Goel VK, Fromknecht SJ, Nishiyama K, et al. The role of lumbar spinal elements in flexion. Spine 1985; 10:516–23.
14. Hindle RJ, Pearcy MJ, Cross A. Mechanical function of the human lumbar interspinous and supraspinous ligaments. J Biomed Eng 1990; 12:340–4.
15. Iguchi T, Kurihara A, Nakayama J, et al. Minimum 10-year outcome
of decompressive laminectomy
for degenerative lumbar spinal stenosis
. Spine 2000; 25:1754–9.
16. Jalovaara P, Lähde S, Iikko E, et al. The significance of residual stenosis after decompression for lumbar spinal stenosis
. Ann Chir Gynaecol 1989; 78:304–8.
17. Johnsson KE, Uden A, Rosen I. The effect of decompression on the natural course of spinal stenosis. A comparison of surgically treated and untreated patients. Spine 1991; 16:615–19.
18. Onel D, Sari H, Dönmez C. Lumbar spinal stenosis
: Clinical/radiological therapeutic evaluation in 145 patients. Conservative treatment or surgical intervention? Spine 1993; 18:291–8.
19. Herno A, Airaksinen O, Saari T, et al. Lumbar spinal stenosis
: A matched-pair study of operated and non-operated patients. Br J Neurosurg 1996; 10:461–5.
20. Amundsen T, Weber H, Nordal HJ, et al. Lumbar spinal stenosis
: Conservative or surgical management? A prospective 10-year study. Spine 2000; 25:1424–36.
21. Atlas SJ, Keller RB, Robson D, et al. Surgical and nonsurgical management of lumbar spinal stenosis
. Spine 2000; 25:556–62.
22. Surin V, Hedelin E, Smith L. Degenerative lumbar spinal stenosis
. Results of operative treatment. Acta Orthop Scand 1982; 53:79–85.
23. Katz JN, Lipson SJ, Larson MG, et al. The outcome
of decompressive laminectomy
for degenerative lumbar stenosis. J Bone Joint Surg [Am] 1991; 73:809–16.
24. Herno A, Airaksinen O, Saari T. Long-term results of surgical treatment of lumbar spinal stenosis
. Spine 1993; 18:1471–4.
25. Katz JN, Lipson SJ, Brick GW, et al. Clinical correlates of patient satisfaction after laminectomy
for degenerative lumbar spinal stenosis
. Spine 1995; 20:1155–60.
26. Jönsson B, Annertz M, Sjöberg C, et al. A prospective and consecutive study of surgically treated lumbar spinal stenosis
. Part II: Five-year follow-up by an independent observer. Spine 1997; 22:2938–44.
27. Guigui P, Benoist M, Delecourt C, et al. Motor deficit in lumbar spinal stenosis
: A retrospective study of a series of 50 patients. J Spinal Disord 1998; 11:283–8.
28. Hurri H, Slätis P, Soini J. Lumbar spinal stenosis
: Assessment of long-term outcome
12 years after operative and conservative treatment. J Spinal Disord 1998; 11:110–15.
29. Katz JN, Stucki G, Lipson SJ, et al. Predictors of surgical outcome
in degenerative lumbar spinal stenosis
. Spine 1999; 24:2229–33.
30. Getty CJM. Lumbar spinal stenosis
. The clinical spectrum and the results of operation. J Bone Joint Surg [Br] 1980; 62:481–5.
31. Nakai O, Ookawa A, Yamaura I. Long-term roentgenographic and functional changes in patients who were treated with wide fenestration for central lumbar stenosis. J Bone Joint Surg [Am] 1991; 73:1184–91.
32. Turner JA, Ersek M, Herron L, et al. Surgery for lumbar spinal stenosis
. Attempted meta-analysis of the literature. Spine 1992; 17:1–8.
33. Postacchini F., Cinotti G, Gumina S, et al. Long-term results of surgery in lumbar stenosis. 8-year review of 64 patients. Acta Orthop Scand
34. Airaksinen O, Herno A, Saari T. Surgical treatment of lumbar spinal stenosis
: Patients' postoperative disability and working capacity. Eur Spine J 1994; 3:261–4.
35. Jönsson B, Strömqvist B. Lumbar spine surgery in the elderly. Complications and surgical results. Spine 1994; 19:1431–5.
36. Herno A, Airaksinen O, Saari T, et al. Surgical results of lumbar spinal stenosis
. A comparison of patients with or without previous back surgery. Spine 1995; 20:964–9.
37. Katz JN, Lipson SJ, Chang LG, et al. Seven- to 10-year outcome
of decompressive surgery for degenerative lumbar spinal stenosis
. Spine 1996; 21:92–8.
38. Niggemeyer O, Strauss JM, Schulitz KP. Comparison of surgical procedures for degenerative lumbar spinal stenosis
: A meta-analysis of the literature from 1975 to 1995. Eur Spine J 1997; 6:423–9.
39. Airaksinen O, Herno A, Turunen V, et al. Surgical outcome
of 438 patients treated surgically for lumbar spinal stenosis
. Spine 1997; 22:2278–82.
40. Johnsson KE, Rosen I, Uden A. The natural course of lumbar spinal stenosis
. Clin Orthop Rel Res 1992; 279:83–6.
41. Katz JN, Lipson SJ, Lew RA, et al. Lumbar laminectomy
alone or with instrumented or noninstrumented arthrodesis in degenerative lumbar spinal stenosis
. Patient selection, costs, and surgical outcomes. Spine 1997; 22:1123–31.
42. Malter AD, McNeney B, Loeser JD, et al. 5-year reoperation rates after different types of lumbar spine surgery. Spine 1998; 23:814–20.
43. Charafeddine H, Gangloff S, Onimus M. L'instabilité post-opératoire après laminoarthrectomie pour sténose lombaire dégénérative. Rev Chir Orthop 1994; 80:379–87.
44. Frazier DD, Lipson SJ, Fossel AH, et al. Associations between spinal deformity and outcomes after decompression for spinal stenosis. Spine 1997; 22:2025–9.
45. Herno A, Airaksinen O, Saari T. Computed tomography after laminectomy
for lumbar spinal stenosis
. Patients' pain patterns, walking capacity, and subjective disability had no correlation with computed tomography findings. Spine 1994; 19:1975–8.
46. Herno A, Partanen K, Talaslahti T, et al. Long-term clinical and magnetic resonance imaging follow-up assessment of patients with lumbar spinal stenosis
. Spine 1999; 24:1533–7.
47. Herno A, Saari T, Suomalainen O, et al. The degree of decompressive relief and its relation to clinical outcome
in patients undergoing surgery for lumbar spinal stenosis
. Spine 1999; 24:1010–14.