Surgery for Spondylolisthesis: RCT Evidence from the Cochrane Review

J. N. Alastair Gibson

Gordon Waddell

Ideally, any surgical intervention should be evidence based, and surgery for spondylolisthesis is no exception. The first, clinical demand is that surgery should be effective, but in these days of finite health care funding, it should also be cost effective. Indeed, a marginally less effective, yet safer or cheaper intervention may be the treatment of choice in some situations. This may be especially relevant to the choice of intervention for spondylolisthesis.

We have recently collated all randomized controlled trials (RCTs) on surgical interventions for the latest update of the Cochrane Review of Degenerative Lumbar Spondylosis (1). From that review, it was possible to extract data pertaining to the treatment of spondylolysis and the various types of spondylolisthesis for this chapter. Critical analysis allows important questions to be addressed with respect to the type of surgery, timing of surgery, and cost.

METHODS

Selection of Trials

All trials published as full articles or abstracts were collected from computer searches of Medline, PubMed, and the Cochrane Controlled Trials Register. Spine, the Journal of Spinal Disorders, and other specialized periodicals were hand-searched and, in addition, surgical colleagues around the world forwarded RCTs known to them. All RCTs were checked by both authors for details of the trial methodology and quality rated. Those RCTs with any reference to spondylolysis or spondylolisthesis were then extracted and all relevant data analyzed for this paper.

Outcome Analysis

Outcome data presented in the trials varied considerably in its content and substance, but most trials contained some relevant results that could be analyzed using the Cochrane review manager PC software (Revman®). Where appropriate, meta-analysis was performed and odds ratios (OR) with 95% confidence limits (CI) calculated. Patient or surgeon ratings of outcome were treated by pooling the data, classifying excellent or good outcomes as “success” and fair or poor as “failure.”

TRIALS

In the 2005 update of the Cochrane Review on lumbar spondylosis, 31 published RCTs are included. Eleven contain data relating to patients with spondylolisthesis. Two of these analyzed outcomes following treatment of an isthmic type spondylolisthesis, while the remainder assessed some aspect of treatment of spinal stenosis in conjunction with degenerative spondylolisthesis. The characteristics of these trials are presented in Table 31.1.

TABLE 31.1. Characteristics of trials

Trial

Methods

Interventions

Participants

Outcomes

Laminotomy versus laminectomy

Postacchini et al. (6)

Alternate assignment

Blinding: nil

Rating: C

Experiment: Multiple laminotomy

Control: Laminectomy

70 patients:

34 male, 36 female
Age 43-79 years

Outcomes at 3.7 years:

Patient’s rating

Surgeon’s rating

Spondylolisthesis progression

Operating time

Blood loss

Arthrodesis versus decompression alone

Herkowitz and Kurz (8)

Alternate assignment

Blinding: nil

Rating: C

Experiment:

Posterolateral arthrodesis and decompression

Control:

Decompression

50 patients:

14 male, 36 female
Age 52-84 years
Loss to follow-up: 0/50

Outcomes at 3 years:

Back pain

Leg pain

Surgeon’s rating

Fusion

Spondylolisthesis progression

Bridwell et al. (7)

Randomization method unknown

Blinding: nil

Rating: B

Experiment:

a. Posterolateral fusion

b. Instrumented posterolateral fusion (Steffee system mostly)

Control:

Decompression

44 patients:

10 male, 34 female
Age 44-79 years
Loss to follow-up: 1/44

Outcomes at 2 years:

Spondylolisthesis progression

Secondary surgery

Walking distance

Fischgrund et al. (10)

Closed envelope allocation

Assessor blinded

Rating: A

Experiment:

Instrumented

posterolateral fusion

(Steffee system)

Control:

Posterolateral fusion

76 patients:

17 male, 59 female
Age 52-86 years
Loss to follow-up: 8/76

Outcomes at 2 years:

Back pain score

Leg pain score

Surgeon’s rating

Fusion

Progression of spondylolisthesis

Kitchel and Matteri (11)

Randomization method unknown

Blinding: nil

Rating: B

Experiment:

Instrumented posterolateral (system unknown) and posterior interbody fusion (autogenous graft)

Control:

Instrumented posterolateral fusion (system unknown)

62 patients: no details

Outcomes at 2 years:

Fusion

Change in Oswestry

Disability Index

Isthmic spondylolisthesis

Carragee (2)

Closed envelope randomization

Blinding: nil

Rating: A

Experiment:

a. Smokers with instrumented arthrodesis (Texas Scottish Rite Hospital System)

b. Nonsmokers with graft alone

Control:

Same groups with no laminectomy

42 patients:

26 male, 16 female
Age 19-51 years
Loss to follow-up: 2/42

Outcomes at 3 years:

Fusion

Patient’s rating

Moller and Hedlund (29,30)

Blindly selected choice of three

Blinding: nil

Rating: A

Experiment:

a. Instrumented posterolateral fusion (Cotrel-Dubousset system)

b. Posterolateral fusion

Control:

Exercise program

111 patients:

57 male, 54 female
Age 18-55 years
Loss to follow-up: 8/114 randomized

Outcomes at 2 years:

Disability rating index

Pain score

Assessor rating

Patient rating

Fusion

Return to work

Spondylolisthesis instrumentation

Zdeblick (16)

Random number generation

Blinding: nil

Rating: A

Experiment:

a. Instrumented posterolateral fusion (Texas Scottish Rite Hospital System)

b. Semirigid instrumented posterolateral fusion (Luque II system)

Control:

Posterolateral fusion

124 patients:

Age 20-80 years
Loss to follow-up: 1/124

Outcomes at 16 months:

Surgeon’s rating

Second procedure

Thomsen et al. (15)

Closed envelope randomization

Assessor blinded

Rating: A

Experiment:

Instrumented posterolateral fusion (Cotrel-Dubousset system)

Control:

Posterolateral fusion

130 patients:

60 male, 69 female
Age 20-67 years
Loss to follow-up: 3/129

Outcomes at 2 years:

Functional scale

Patient’s rating

Fusion

Second procedure

France et al. (14)

Randomization method unknown

Blinding: nil

Rating: B

Experiment:

Instrumented posterolateral fusion (Steffee system)

Control:

Posterolateral fusion

83 patients:

58 male, 25 female
Age 19-76 years
Loss to follow-up: 12/83

Outcomes at 2 years:

Back pain scale

Patient’s rating

Fusion

Christensen et al. (17)

Closed envelope randomization

Independent radiologist

Rating: A

Experiment:

Circumferential fusion with anterior interbody fusion (Brantigan cage) and posterior instrumentation (Cotrel-Dubousset system or transarticular screws)

Control:

Instrumented posterolateral fusion (Cotrel-Dubousset system)

148 patients:

88 male, 58 female
Age 20-65 years
Loss to follow-up: 9/146

Outcomes at 2 years:

Dallas pain score

Low back rating

Work status

Note: Allocation concealment was rated into three grades: (A) clearly yes, some form of centralized randomization scheme or assignment; (B) unclear, assignment envelopes, a list, or table, evidence of possible randomization failure, or those trials stated to be random but with no description of the method; (C) clearly no, including quasi-randomization where patients were allocated by alternation, case number, or date of birth in which the method of allocation was transparent before assignment to a given group.

Only gold members can continue reading. Log In or Register to continue