Decompression alone versus decompression with limited fusion for treatment of degenerative lumbar scoliosis in the elderly patient.

STUDY
DESIGN: Retrospective cohort study.
OBJECTIVE: To analyze the surgical results of a group of patients older than 65 years treated for mild degenerative lumbar scoliosis (<30°) with stenosis, treated with decompression alone or decompression and limited fusion.
METHODS: We evaluated 55 patients, all older than 65 years from our prospectively collected database with mild degenerative scoliosis (<30°) and stenosis who underwent surgery. Laminectomy alone was performed in 16 patients, and laminectomy and limited fusion in 39 patients. Mean follow-up was 4.6 years in the decompression group and 5.0 years in the fusion group. Clinical results were graded by patients' self-reported satisfaction and length of symptom-free period to recurrence.
RESULTS: In the decompression alone group, 6 (37%) of 16 patients developed recurrent stenosis at the previously decompressed level and five developed recurrence within 6 months postoperatively versus the decompression and fusion group where 3 (8%) of 39 (P = .0476) developed symptomatic stenosis supra adjacent to the fusion. Of 16 patients in the decompression alone group, 12 (75%) had recurrence of symptoms by the 5-year follow-up period versus only 14 (36%) patients in the decompression and fusion group (P = .016). Adjacent segment degenerative changes were common in the fusion group, but only 7% developed symptomatic stenosis.
CONCLUSIONS: Decompression with limited fusion prevents early return of stenotic symptoms compared with decompression alone in the setting of mild degenerative scoliosis (<30°) and symptomatic stenosis in patients 65 years and older. [Table: see text] The definiton of the different classes of evidence is available on page 67.

Study Rationale and Context

The surgical treatment of degenerative lumbar scoliosis with stenosis (DLS) in the elderly is challenging.1,2,3,4,5 The goal is to reduce surgical morbidity and achieve good, long-standing results.6 In patients who are not considered candidates for a long thoracic and lumbar fusion, decompression alone and decompression with limited fusion are the main treatment options. There are few reports evaluating older patients with DLS.

Objective

To compare the surgical treatment results of decompression only to decompression and fusion in patients older than 65 years with mild degenerative scoliosis (<30°) and symptomatic stenosis consisting of radiculopathy or neurogenic claudication.

Patient sampling and selection.

Methods

Retrospective cohort study.

Patients older than 65 years with degenerative scoliosis

Degenerative scoliosis less than 30°

Lumbar stenosis with radiculopathy or symptoms of neurogenic claudication refractory to non-surgical treatment

Patients with a minimum 24-month follow-up

Patients younger than 65 years

Degenerative scoliosis greater than 30°

Patients without radiculopathy or neurogenic claudication symptoms

Fifty-five consecutive patients with DLS and stenosis who underwent surgery by two spine surgeons at one institution from 1996–2003 were retrospectively reviewed radiographically and clinically. All patients were treated during the same period. The standard of care at our institution for the treatment of degenerative scoliosis of less than 30° with stenosis is laminotomy/laminectomy without fusion, or laminotomy/laminectomy with limited fusion at the levels of decompression only.

The technique for the decompression only (group D) consisted of a standard bilateral laminotomy and foraminotomy or hemilaminectomy with an effort to preserve at least 50% of the facet bilaterally.

The technique for decompression and fusion (group DF) included a similar technique for decompression as previously described, as well as the addition of an instrumented fusion. The instrumented fusion consisted of the placement of pedicle screw instrumentation at the level or levels decompressed, followed by the placement of a connecting rod. The rod was bent to match the curve on the table and no active corrective maneuvers were used. Iliac crest and local autograft was used for the posterior intertransverse arthrodesis.

The following outcomes were measured:

This data was gathered from the medical record at follow-up intervals (6 weeks, 3 months, 6 months, 12 months, and 24 months). If the patients began to have recurrent symptoms before the routine follow-up period, and the specific time was noted, this was recorded.

(Table 1: This non-validated clinical assessment was developed by the authors to evaluate the outcomes. The clinical assessment category was determined from written information contained in the medical record. The clinical assessment category was determined by an independent surgeon who was not one of the two senior treating surgeons. The reviewer was not blinded to the treatment. An ‘excellent’ outcome was identified as a patient with complete relief of his/her radicular or neurogenic symptoms. A good result was defined when symptoms were improved without recurrent radicular symptoms, but not overall completely symptom free. A fair result was defined as one with mild symptoms of recurrence similar to preoperative symptoms, but improved compared with preoperative condition. A poor result was recurrence of preoperative symptoms, little improvement, or symptoms that were unchanged.

Table 1

Clinical results at 2-year follow-up.

		Decompression alone (n = 16)No. (%)	Decompression/limited fusion (n = 39)No. (%)	P*
Excellent	Symptom relief	9 (56)	28 (72)	.16
Good	Minor residual symptomsNo recurrence, better than preoperative	1 (6)	4 (10)
Fair	Recurrence, better than preoperative	0	4 (10)	.07
Poor	Recurrence, unchanged or worse than preoperative	6 (38)	3 (8)

Comparisons were made using a Fisher exact test. Two groups were compared: excellent/good; fair/poor.

The global clinical assessment and radiographic Cobb angle measurements were recorded at the 2-year period for all patients and at the last follow-up for patients with 5-year follow-up.

A diagnosis of radiographic adjacent segment degeneration (ASD) was assigned when there was >75% disc space narrowing, large spur formation, olisthesis or translation ≥ 5 mm at the disc level adjacent to the fused segments8 or when there were symptoms of spinal stenosis (neurogenic claudication or radiculopathy) referable to the adjacent segment and confirmed by selective nerve root injections.

Unpaired t tests were performed to analyze the demographic differences between patients who underwent a decompression alone and those who underwent a decompression and limited fusion.

Unpaired t tests were also used to evaluate the differences in length of symptom-free time.

Fisher exact test was used to determine the differences in the global assessment and the presence of ASD.

A paired t test was used to evaluate the differences in radiographic Cobb angles within each subject.

Mean age was 75 years in the decompression group (D) and 66 years in the limited fusion group (DF) (P = .01). Mean follow-up was 4.6 years in the decompression group and 4.6 years in the fusion group (Table 2).

Table 2

Demographics and radiographic data.

	Decompression alone (n = 16)	Decompression/limited fusion (n = 39)	P
Average age, y	75 ± 6	66 ± 7	.01*
Average follow-up, y	4.6 (± 2.7)	4.6 (± 2.4)	.39*
Average preoperative lumbar scoliosis, °	16 ± 6	22 ± 8	.23*
Average postoperative correction, °	3 ± 4	1 ± 8	.5*
Levels treated† 1 2 3 4	5 9 1 1	1227	.74†

Unpaired t test.

Fisher exact test.

Hundred percent of patients had follow-up at the 2-year follow-up.

Twenty-five (45%) of 55 patients had 5-year follow-up.

Overall, satisfactory (good-to-excellent) results were found in 10 (63%) of 16 in Group D and 32 (82%) of 39 in Group DF at 2-year follow-up (P = .16).

In Group D, 10 (63%) of 16 patients remained symptom free at 2-year follow-up, and in Group DF, 36 (92%) of 39 patients remained asymptomatic at 2 years (P = .018) (Table 3).

Table 3

Time interval of recurrent leg pain.

Recurrent spinal stenosis	Decompression alone n = 16 (%)	Decompression/limited fusion n = 39 (%)	P*
Within first 6 mo	5 (31)	0 (0)	.001
Between 6 mo and 2 y	1 (9)	3 (7)	1.00
Within 2-y follow-up	6 (37)	3 (8)	.018

Comparisons were made using Fisher exact test.

In Group D, 4 (25%) of 16 had evidence of ASD while in Group DF 20 (51%) of 39 patients had evidence of ASD (P = .13).

In Group D, the mean preoperative Cobb angle was 16°, and 18° at last follow-up for a mean change of 2° (range, 0–8°), and in Group DF, the mean Cobb angle did not change from preoperative to last follow-up at a value of 22° for a mean change of 0° (range, 0–17°) (P = .17).

There were 25 patients overall (47%) with at least 5-year follow-up, 8 (50%) in Group D and 17 (43%) in Group DF. At 5-year follow-up, 4 (50%) of 8 patients in Group D had satisfactory (good) results and 3 (37%) of 8 had evidence of ASD. In Group DF, 10 (59%) of 17 had satisfactory (good) results (P = 1.0), and 14 (82%) of 17 showed disc space collapse at the supra-adjacent segment.

Of 16 patients in the decompression alone group, 12 (75%) had recurrence of symptoms by the 5-year follow-up versus only 14 (36%) in the decompression and fusion group (P = .016).

Complications are listed in Table 4.

Table 4

Complications.

	Decompression alone n = 16 (%)	Decompression/limited fusion n = 39 (%)	P†
Deep vein thrombosis	0 (0)	1 (2)	1.00
Dural tear	1 (6)	1 (2)	.50
Deep wound infection	0 (0)	2 (5)	1.00
Reoperation – Decompression – Fusion	1 (6)3 (18)	2* (5)	.053

Two additional patients are candidates for revision surgery.

Comparisons were made using Fisher exact test.

A significant number of patients who had decompression without fusion had recurrent symptoms (37%) at the same level within the 2-year follow-up versus only 8% in the decompression and fusion group (P = .018) (Table 3).

Decompression and fusion had more sustainable results in regard to recurrence of symptoms at 2 years and at final follow-up compared with decompression alone.

In this consecutive series of patients, there was selection bias in determining the type of surgical treatment offered to patients with DLS and stenosis. Patients who underwent a decompression alone were significantly older in age compared with those with decompression and limited fusion.

Progression of scoliosis was not seen in our patient population regardless of the type of treatment. The treatment of symptomatic stenosis in well-selected patients with mild scoliosis < 30° with decompression or decompression and limited fusion does not appear to destabilize the spine enough to cause curve progression.

With further follow-up (5 years), there was still no evidence of curve progression in either group; however, the fusion group had a higher incidence of ASD at the supra-adjacent level,7,8 and in general, the results of both treatment methods deteriorated with time.

This is a difficult patient population to treat and there are no clear standards.1,2,3,4,5 As this series shows, we tend to treat older patients with the less invasive procedure but the symptoms often quickly recur leaving doubt to its ultimate effectiveness. If we add fusion, we seem to lengthen the symptom-free period.

The surgeons in this series did not have a formal algorithm that guided their treatment. However, they are both experienced spinal deformity surgeons with an awareness of the potential for curve progression with decompression alone. While age was significantly different between the treatment groups, this was only recognized after the statistical analysis. In general, patients at our institution who undergo a decompression alone will have large stabilizing osteophytes, severe disc space collapse, and mild scoliotic curves < 30°.

There are many confounding factors in this study that are inherent to its retrospective nature and may have impacted the results. The fact that both senior surgeons are experienced in treating deformity and may be better at selecting patients who are more appropriate for decompression alone may have had an impact on the results.

While there was no significant curve progression seen in this study, there was significant recurrence of stenotic symptoms in the decompression alone group. So, it would appear that the surgeons were able to accurately assess patients whose curves were stable enough for decompression alone, but were not able to predict in whom symptoms might recur. It appears that symptoms can recur without noticeable curve progression. One can speculate that more subtle changes in rotation and translation may be occurring that are not detected in standard Cobb measurements.

As this study shows, there is inherent bias toward treating older patients with the least invasive procedure (decompression alone) but if the goal is to provide the treatment that gives the longest and most sustainable results, less in these cases may be more. Adding a limited fusion to the decompressed segments seems to improve the length of symptom-free time.

There are many limitations to this study. It is a retrospective case series with many inherent treatment biases and confounding factors that may have affected the results. The smaller number of patients included for comparison that underwent a decompression alone and the higher loss to follow-up in this group (30%) versus the decompression and fusion group (7%) is a limitation. Another weakness is the age differences between the groups. The age difference may account for the differences in outcomes. The outcomes selected were also non-validated, and may not be reliable. This study may also represent a best-case scenario of patients with degenerative scoliosis being treated with a decompression without fusion. This study was performed at a tertiary spine center with extensive experience in treating adult deformity. The biases in clinical judgment and decisions may be a factor in our findings.

Discussion

Limiting a fusion to the decompressed regions only in patients with mild degenerative scoliosis, and stenosis in patients older than 65 years appears to stabilize the decompressed segments enough to prevent early recurrence of stenosis symptoms and curve progression.

Decompression alone led to a higher rate of recurrence of stenotic symptoms.

Age may play a role in the differences noted in our study groups.

Summary and Conclusions

Decompression and limited fusion of the decompressed segments in the setting of mild degenerative scoliosis (<30°) appears to be a more effective treatment for patients 65 years or older with symptoms of stenosis.

Final class of evidence–prognosis
Study design
RCT
Cohort	•
Case control
Case series
Methods
Concealed allocation (RCT)
Intention to treat (RCT)
Blinded/independent evaluation of primary outcome
F/U ≥ 85%	•
Adequate sample size
Control for confounding
Overall class of evidence	III