Surgical treatment for unstable low-grade isthmic spondylolisthesis in adults: a prospective controlled study of posterior instrumented fusion compared with combined anterior-posterior fusion

The Spine Journal 6 (2006) 606–614

2006 Outstanding Paper Award: Surgical Science

Surgical treatment for unstable low-grade isthmic spondylolisthesis in adults: a prospective controlled study of posterior instrumented fusion compared with combined anterior-posterior fusion Justin Swan, MD, Eric Hurwitz, PhD, Farbod Malek, MD, Erica van den Haak, BS, Ivan Cheng, MD, Todd Alamin, MD, Eugene Carragee, MD* Department of Orthopaedic Surgery, Spine Surgery Section, Stanford University Medical Center, 300 Pasteur Drive, Stanford, CA 94305-5326, USA Received 8 January 2006; accepted 22 February 2006

Abstract

BACKGROUND CONTEXT: The surgical treatment for low-grade isthmic spondylolisthesis in adults with intractable lumbar pain is usually spinal fusion. It has been postulated that anterior column reconstruction may be relatively advantageous in those patients with unstable slips. PURPOSE: To compare the early and medium term treatment efficacy of two common fusion techniques in isthmic spondylolisthesis. STUDY DESIGN/SETTING: Prospective controlled trial comparing single-level posterior-lateral instrumented fusion with combined anterior and posterior-lateral instrumented fusion in sequential matched cohorts of patients with radiographically unstable isthmic spondylolisthesis. OUTCOME MEASURES: Primary outcome measure of success was an Oswestry Disability Index (ODI)#20. Secondary outcome measures included patient determined minimum-acceptable outcome on four questionnaires: pain intensity (visual analog scale), ODI, medication intake, and work status. Radiographic outcome of fusion was determined by radiographic union and motion on flexion/extension X-rays. Risk ratios (RRs) and 95% confidence intervals (CIs) were calculated for primary outcome of success for combined fusion compared with posterior fusion. METHODS: The study was conducted over a 6-year period. The first cohort of 50 consecutive patients was treated with a single-level instrumented posterior-lateral fusion; the second sequential cohort was treated with an anterior interbody fusion and the same posterior operation. Observations were made at baseline, 6 months, 1 year, and 2 years after surgery. Final radiographic assessment was made at 2 years after surgery. RESULTS: Baseline demographic and clinical factors were well-matched in the two cohorts. At 2 years, 46 posterior-only fusion subjects and 47 combined fusion subjects completed the full followup regimen. Outcomes were better by all measures at 6 months and 12 months in the anterior-posterior cohort. Comparing the primary outcome measure (ODI outcome#20) in the posterior versus the combined groups, success was achieved at 6 months in 11 versus 30 (RR52.67, 95% CI 1.53, 4.67; p5.0001); at 1 year, 20 versus 34 (RR51.66, 95% CI 1.14, 2.42; p!.005); and at 2 years, 29 versus 36 subjects (RR51.21, 95% CI 0.93, 1.59; p5.14). At 6 months, 13 posterior-only and 25 combined group subjects had returned to work (RR 1.88, 95% CI 1.10, 3.21; p5.01). More patients achieved their preoperatively determined minimum-acceptable outcome at each time point. There were three nonunions in the posterior-alone cohort and one in the combined group. Serious complications and reoperations were similar in both groups. CONCLUSION: Outcomes up to 2 years were superior by clinically important differences after a combined anterior-posterior operation compared with posterior-alone surgery for unstable spondylolisthesis; however, between-group differences attenuated appreciably after 6 months. The apparent clinical and occupational benefits of combined fusion should be considered along with

FDA device/drug status: not applicable. Nothing of value received from a commercial entity related to this manuscript. 1529-9430/06/$ – see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2006.02.032

* Corresponding author. Stanford University Medical Center, Department of Orthopaedic Surgery, Spinal Surgery Section, 300 Pasteur Drive, Stanford, CA 94305-5326. Tel.: (650) 725-6797; fax: (650) 723-9805. E-mail address: [email protected] (E. Carragee)

J. Swan et al. / The Spine Journal 6 (2006) 606–614

607

possible increases in minor complications and procedure-related costs. Ó 2006 Elsevier Inc. All rights reserved. Keywords:

Isthmic spondylolisthesis; Degenerative spondylolisthesis; Posterolateral fusion; Anterior-posterior fusion; Combined fusion; Instrumented fusion; Unstable spondylolisthesis

Introduction The majority of the persons with isthmic spondylolisthesis are not clinically symptomatic. The initial treatment in most symptomatic patients is nonsurgical supportive care [1]. The mainstay of surgical treatment for adult patients with persistent symptomatic isthmic spondylolisthesis is usually fusion [2], with or without decompression [3,4]. Fusion techniques considered for the treatment of this deformity include posterolateral intertransverse process fusion, anterior, posterior, or transforaminal interbody fusion, and combined anterior and posterior fusion [2,5]. Much has been written about the theoretical advantages of each approach; however, no clearly optimal approach has been established to date. The current controlled trial is a follow-on study to a randomized control trial of surgical treatment of isthmic spondylolisthesis performed at our center from 1993 to 1995 [3]. In that trial, subjects treated with posterior-lateral fusion did not appear to benefit from concurrent laminectomy when there was no serious neurologic loss preoperatively. Concurrent review, however, identified relatively poorer outcomes in subjects with more mobile slips compared with subjects with slips that were either stable over time or showed little motion on dynamic radiographs. Subgroup analysis suggested better results in subjects treated with instrumented fusions when dynamic radiographs revealed clear motion. On the other hand, subjects with little motion seemed to do as well or better without transpedicular instrumentation. Similar conflicting reports on the efficacy of instrumentation in treating isthmic spondylolisthesis have been reported by others [2,6–14]. Many slips appear to move little or not at all on dynamic testing [15,16]. It is not clear that more aggressive fusion techniques (eg, instrumentation, interbody grafting) are appropriate in cases with minimal motion when fusion may otherwise be indicated. Radiographic instability certainly exists but is neither well-defined nor commonly appreciated in patients with symptomatic spondylolisthesis [15,17–19]. Progression of the degree of anterolisthesis over time is one measure of instability [17]. Another is relatively increased translation or angular motion seen either on flexion and extension X-rays or upright and prone comparisons [19,20]. We have postulated that instrumentation or anterior column fusion techniques may be most appropriate in these subjects with some degree of instability, but the advantages of one compared with another have not been proven. To explore this issue, we conducted a prospective clinical trial comparing posterior instrumented fusion alone

with a combined anterior and posterior instrumented fusion in the subgroup of patients with isthmic spondylolisthesis who had radiographic evidence of relative instability.

Materials and methods Study design This is a sequential cohort study of two groups, with 50 subjects in each treatment arm. Fifty consecutive subjects were treated with instrumented posterior spinal fusion on an established evaluation and follow-up protocol (1995– 1997). An equal number of subjects were then treated with a combined anterior and posterior spinal fusion using the same protocol (1998–2001). The study was approved by the Human Subjects Research Committee at Stanford University and followed Department of Health and Human Services guidelines. Enrollment criteria. Consecutive patients (1995–2001) with intractable low back pain (with or without leg pain) and radiographically confirmed, unstable Grade I or II isthmic spondylolisthesis of either the L5–S1 or L4–L5 anatomic segments were considered for enrollment. Unstable spondylolisthesis was defined in this study as follows: documented slip progression (3 mm or one Meyerding grade) under observation in the 2 years before surgery; or $3 mm translation and/or $22 of angulation as seen on standing flexion-extension or recumbent (prone) lateral radiographs. Patients were excluded if the preoperative evaluation found: greater than trace motor weakness; retrolisthesis, disc protrusion, painful disc injection, or instability of an adjacent segment; positive straight leg raising sign; metabolic bone disease; previous spinal surgery, other lumbar deformity (O15 scoliosis), or fracture; inflammatory arthritis/autoimmune disease. Enrollment was limited to subjects undergoing single-level fusions. Posterior instrumented fusion. The posterior instrumentation and fusion procedure alone was performed in the first cohort. A midline incision, followed by a small Wiltse paraspinal interval exposure, was used in each patient. An operating microscope was used to assist dissection to expose the transverse processes, pars, and facet region, which were thoroughly decorticated. No decompression was performed. Bilateral pedicle screw instrumentation (Variable Angle Screw, Synthes, Paoli, PA) was applied. Transpedicular screw insertion was directed by fluoroscopic imaging. Autograft was harvested, through the same incision, from the inner table of the iliac crest and grafted in each patient.

608

J. Swan et al. / The Spine Journal 6 (2006) 606–614

Combined anterior and posterior fusion. The first stage was an anterior mini-laparotomy (retroperitoneal approach) to expose the affected disc. The anterior annulus was exposed by gentle sweeping of the autonomic plexus and superficial membranous tissue from the anterior spine using dental roll dissection. No electrocautery was used in males along the ventral spine. An anterior subtotal discectomy including the posterior annulus using an operating microscope was performed followed by end plate decortication. A lordotic shaped femoral ring allograft and buttress screw were placed with segmental distraction and reduction of the slip angle (Synthes, Paoli, CA). Local and allograft bone was packed within and around the structural graft. The second stage involved an instrumented posterior-lateral and intertransverse fusion identical to the first cohort. Baseline data Preoperative baseline measures were performed on all patients, including the visual analog scale (VAS) for the severity of back and leg pain, the American Academy of Orthopaedic Surgeons Modems instrument [21], the Oswestry Disability Index (ODI) [22], and psychometric testing using the Modified Zung Depressive Index and the Modified Somatic Perception Questionnaire scores [23]. Each patient also recorded, on a standard questionnaire, a pain drawing, demographic information, history of pain medication usage, preoperative symptom duration, and the amount of preoperative work loss. During the preoperative evaluation, some patients had ancillary evaluations, including electromyography, spondylolysis anesthetic blockade, provocative discography, facet joint or medial branch blocks, or sacroiliac joint blocks in an effort to evaluate possible root involvement or other sites that may be causing clinically significant pain. All subjects meeting the entry criteria above were examined with a high-quality (1.5 Tesla) lumbar spine magnetic resonance imaging, and scans were graded by experienced examiners in a blinded manner by a set protocol. Details of the magnetic resonance imaging protocol have been previously described. Grading of disc degeneration [24], annular disruption [25,26], and end plate status [27] followed previously reported methodology. Details of blood loss, operative time, transfusion during hospitalization, hospital stay, and perioperative complications were made on standardized forms by the spine surgery resident or fellow at the time of discharge. Follow-up assessment Subjects were followed at intervals for 2 years after the index surgery (6 months, 12 months, and 24 months) by independent research assistants at the spine center’s research section (TVT, BY, NVG). The research assistant made mailings and telephone interviews monthly based on a coded list of all surgical patients (of any type) having

had their index operation 6 months, 12 months, or 24 months previously. The follow-up interviews were made in a blinded manner to the diagnosis and procedure performed. These evaluations were made independently from the usual outpatient clinic follow-up examinations. Standard mailing of outcomes metrics included the VAS for back and leg pain, ODI, American Academy of Orthopaedic Surgeons (Modems Lumbar Questionnaire), analgesic usage (none, occasional non-narcotic, daily non-narcotic, occasional narcotic, daily narcotic analgesics), work status (none, part-time, full-time; same, increased, or decreased intensity compared with preoperative work), reoperation, and surgical complications. Telephone interviews made in the same month included a scripted interview concerning interval ancillary treatment, classification of work type, and full-time/part-time status, as well as neurologic symptoms including bowel, bladder, and sexual function. Telephone interview results were recorded on standardized and coded forms. All follow-up information was then entered into the computer database linking coded data to baseline clinical data and operative details. Until the study’s completion, coded follow-up data could not be matched to baseline clinical, demographic, or radiologic data. Protocol radiographs were obtained approximately within 2 months after surgery. Other radiographs were taken as dictated by clinical routine or need. A full final radiographic evaluation on protocol was performed at 2 years postoperatively with anterior and posterior radiographs and standing lateral flexion and extension radiographs. Fusion was judged complete if there was clear bridging trabeculation and no motion on flexion/extension films. Uncertain cases were evaluated with computed tomography if the patient was significantly symptomatic. The baseline disc height, listhesis, and slip angle were measured and compared with the 2-month standing radiograph. To correct for variations in X-ray technique and magnification, the disc height and listhesis distances are represented as percent of the anterior/posterior length of inferior end plate of the cephalad vertebrae [28]. Criteria for surgical success and minimal acceptable outcomes The primary criteria for success at each survey point to 2-year follow-up was defined as an ODI#20 [29]. This was based on a previously defined ‘‘expected’’ result for spondylolisthesis patients about to have spinal fusion. There were two secondary outcomes considered: a minimum-acceptable outcome achievement and return to fulltime work status. Full-time work had been shown in our center to be an expected outcome of more than 90% of working-age subjects having lumbar fusion surgery [29] and was anticipated to be a crucial outcome measure in our subjects. A method to prospectively determine a subject’s ‘‘minimal-acceptable’’ outcomes has been previously

J. Swan et al. / The Spine Journal 6 (2006) 606–614

described for spinal fusion patients from our institution [29]. The minimum-acceptable outcome for each subject was defined preoperatively as the absolute worst outcome for each outcome question-item, which if the patient knew would not be achieved by 2 years after surgery, then they would not agree to have the procedure. Patients in the week before surgery completed standard questionnaires indicating the minimal-acceptable outcome for the pain intensity VAS, ODI, work status, and medication usage. At the three follow-up interval examinations, each subject’s interval status was compared against achieving their own preoperatively determined minimally acceptable scores for all four outcome dimensions. Statistical analyses and statistical power At the time of the study design, literature on single-level fusions in large series surgery suggested approximately an 80% (6SD 14) clinical success rate. To detect with 80% power at the 95% confidence level (alpha50.05) a moderate (20%) difference in effect on clinical success, at least 25 subjects in each cohort would be needed. However, for a small effect (10%), 50 subjects in each group would be needed. Descriptive statistics were used to summarize the patient characteristics measured at baseline for each treatment group. Groups were compared on known prognostic factors to determine the need for multivariate statistical modeling to control confounding. Given that the groups were found comparable, only unadjusted effect estimates are presented. Risk ratios and 95% confidence intervals were calculated for all outcomes at each follow-up point, comparing combined fusion with posterior fusion.

Results During the study period, 347 subjects were considered and agreed to operative treatment for isthmic spondylolisthesis at the Stanford University Spine Clinic. Of these, 102 were excluded from this trial due to co-morbid conditions or multilevel disease, 122 showed little or no motion on radiographs, 20 subjects refused fusion and elected for root decompression alone, and 2 men who would have been in the combined approach group declined the anterior surgery because of concerns for retrograde ejaculation. One subject designated for the posterior-alone group recovered sufficiently between consenting to the study and the surgery date that she did not have any operation. Of the 100 subjects included in the study, 19 had documented progressive listhesis within 1 year of surgery, 11 within 2 years of surgery, 46 had instability on flexion/extension X-rays or recumbent–standing X-rays, and 33 had instability on more than one mode of evaluation. Seven subjects were lost to follow-up or declined to complete forms after entry to the study. The baseline

609

demographic, imaging, and clinical values for both treatment groups and drop-out subjects are shown in Tables 1 and 2. There were no important differences in any baseline characteristics, and the two groups appeared well-matched. Subjects lost to follow-up did not appear to be different in either demographic or clinical characteristics at baseline. Possibly related to the instability detected, our subjects had more cephalad positioned lysis and listhesis than usually seen. There were 28 subjects in the combined surgery group and 30 in the posterior-only group with either L4–L5 spondylolistheses or an L5–S1 listhesis above the intercrestal line. The surgical data are presented in Table 3. Operative time for the combined surgery was 1 hour longer than the posterior-alone procedure. Blood loss and hospital stay were similar. Clinical outcomes Clinical outcomes data for pain intensity and function are given in Table 4. At 6 months and 1 year, the mean pain intensity (VAS), ODI, medication, and occupational outcomes in the combined fusion group were appreciably better than outcomes in the posterior-only fusion. Medication and work status at each follow-up point are given in Table 5. Again, the combined surgery group appeared to be doing better at the 6- and 12-month evaluations. The satisfactory primary and secondary outcome measures are given in Table 6. The primary outcome (ODI#20) was achieved by more subjects in the combined surgery group compared with the posterior-alone at each followup interval. Full-time work was achieved again, earlier and more frequently in the combined surgery group. Minimum-acceptable outcomes were achieved by most subjects in both groups by 2-year follow-up (Table 6), but again proportionally more subjects in the combined surgery group achieved these goals by 6, 12, and 24 months after surgery. At 2 years more subjects met their minimal outcomes goals after combined fusion (89%) than after posterior fusion alone (70%), risk ratio 1.28 (95% confidence interval 1.04, 1.59). Radiographic outcomes Listhesis magnitude, slip angle, and disc height were all improved on early postoperative radiographs (Table 2). This correction was lost in the posterior-only group but largely maintained in the combined surgery group. There were four radiographic nonunions (one in the combined groups and three in the posterior-only group). Complications and reoperations Most complications reported were minor but minor complications were more frequent after combined surgery (Table 7). In an interesting observation, three subjects reported anterior thigh pain after the combined procedure.

610

J. Swan et al. / The Spine Journal 6 (2006) 606–614

Table 1 Clinical and baseline demographic data

Number of patients Male Female Average age Workers Compensation Smokers Chronic pain diagnosis Preoperative symptoms (months) Average back pain (0–10) Average leg pain (0–10) Average Oswestry Disability Index (0–100) 0–10 10–20 20–40 40–60 O60 AAOS (0–100) MSPQ (1–11) ZUNG FABQ Work status Disabled Light Medium Heavy Working Light Medium Heavy Unemployed/Retired

Combined AP

Posterior-lateral alone

Dropout

47 22 25 42 9 9 6 25629.9 6.7 3.9 53616 0 3 6 28 10 62622 4.964.7 14611.7 14.266.8

46 23 23 43 10 7 4 33613 6.6 4.0 50621 0 3 7 27 9 60623 4.464.2 11.666 13.165.5

3 AP 1 2 41

4 PL 2 3 40

1 0

2 0

6

5.6

18 4 6 8 27 9 9 11 2

16 4 8 4 29 9 10 10 1

45

50

3

4

62 8 18 12.5

60 7.2 13 14

1 0 1 0 2 0 2 0 0

1 0 1 0 3 1 2 0 0

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone; AAOS5American Academy of Orthopaedic Surgeons Modems Lumbar Questionnaire; MSPQ5Modified Somatic Perception Questionnaire; ZUNG5Zung Depression Scale; FABQ5Fear Avoidance Behaviour Questionnaire.

Two occurred after L4–L5 fusion, and all had Grade II slips with Grade V disc degeneration (complete collapse) before reduction and restoration of disc height. All resolved by 3 weeks postoperatively. One of these subjects had minor anterior knee and medial calf numbness long-term, which was not bothersome. Clinically significant problems usually associated with the anterior approach were seen in four subjects (8%): iliac vein lacerations (two repaired without significant blood loss, one repaired but required transfusion); retrograde ejaculation (one, in a diabetic with erectile dysfunction preoperatively, resolved in 6 months). One subject in each group had an exploration of the pedicle screw placement. One was found to be broached medially and was repositioned. Another was found to be completely within the L5 pedicle but was removed, leaving unilateral instrumentation. Sciatica resolved in both cases. There were three nonunions (two smokers) in the posterior-alone group: one had satisfactory outcomes despite the nonunion (VAS52; ODI518; no narcotics and was retired before surgery); another was somewhat symptomatic (VAS 4, ODI 24) but took no medications and was working fulltime in his regular job; the third was severely symptomatic at 1 year and was revised with an anterior lumbar interbody

fusion and screw replacement with larger diameter pedicle screws. This patient had excellent results at 2 years (VAS51, ODI 12 and return to heavy labor). The one nonunion in the combined surgery group (smoker) was doing well (VAS 2, ODI 10, taking daily nonsteroidal anti-inflammatory drugs, working part-time with restricted duties) but had a radiolucent line at the cephalad margin of the allograft with no appreciable motion. Discussion This is the first prospective, controlled trial of surgical treatment strategies focusing exclusively on the patient with low-grade isthmic spondylolisthesis and some degree of instability at the listhetic segment. The results showed clear and clinically significant advantages from anterior column stabilization in both clinical outcome and maintenance of radiographic correction. Previous literature to date has shown conflicting results comparing various treatment strategies in isthmic spondylolisthesis. Moller and Hedlund, in a randomized control trial (n574) of fusion with or without pedicle screw instrumentation, did not find improved outcomes in the instrumented

J. Swan et al. / The Spine Journal 6 (2006) 606–614 Table 2 Preoperative and postoperative imaging data

Table 3 Operative data

PosteriorCombined lateral AP Alone Baseline MRI findings at operative level L4/L5 Disc degeneration grade* 1–2 3–4 5 Annular fissure HIZ End plate changes L5/S1 Disc degeneration grade* 1–2 3–4 5 Annular fissure HIZ End plate changes Radiographic data Anterolisthesis (mm) Baseline 2 months 24 months Disc height (mm) Baseline 2 months 24 months Slip angle (degrees)y Baseline 2 months 24 months

611

Dropout AP

AL

14

12

1

1

0 4 10 3 1 9 33

0 3 9 2 1 10 34

0 0 1 1 0 1 2

0 1 0 0 0 0 3

2 11 19 7 3 12

0 16 18 8 4 10

0 2 0 1 1 1

0 1 1 1 1 1

p value 20.9612.1 21.269.9 8.766.9 16.568.2 9.966.7 19.567.2

.67 .04 .001

17.366.7 26.763.5 24.065.9

.77 .08 .01

16.967.5 22.766.5 18.168.0

18.1611 19.269 27.2613 22.468 25.269 20.2612

.81 .07 .03

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone; HIZ5high intensity zone * MRI disc grading system; Boos et al [24]. y Negative value indicates kyhosis.

group, and possible worse outcomes in that group when the listhesis was Grade II [12]. Others have reported better results with instrumentation [6]. It is not known what proportion of each cohort in either study had stable compared with unstable slips. Recently, Kwon et al. performed a systematic review of outcomes for adult patients undergoing surgery

Time (minutes) EBL (mL) Average hospital stay Transfusion requirement (units/patient)

Combined AP

Posterior-lateral alone p value

244 (195–330) 355 (200–1100) 4.6 (3–7) 0.14

178 (105–245) 310 (100–800) 4.1 (3–8) 0.16

.001 .15 .67 .85

AP5combined anterior and posterior fusion; EBL5estimated blood loss.

for isthmic spondylolisthesis and concluded that a combined anterior-posterior approach may more reliably achieve fusion and a successful clinical outcome [2]. However, they also noted the series reviewed were primarily retrospective, were not consistent in this outcome, and included heterogeneous patient groups with respect to indications for surgery, levels fused, and methods of evaluating outcome. In fact, the two controlled studies with apparently the best-matched treatment groups did not find any advantage in clinical outcome with anterior column arthrodesis. Suk et al. [14] reported on 56 patients treated with pedicle screw fixation and either posterior alone or anterior posterior fusion, although the method of assignment to each treatment group was not described. The disc height and slip angle reduction appeared to be better maintained in the combined surgery group. These authors did not observe a difference in clinical outcomes, but no data were given for functional outcomes, medication usage, or work status. Madan and Boeree [11] assigned subjects by a coin toss at surgery to posterolateral instrumented fusion with or without posterior lumbar interbody fusion but reported better clinical outcomes in patients treated with posterolateral fusion alone, although the fusion rate appeared higher in the posterior lumbar interbody fusion group. These authors hypothesized that the root manipulation required for the posterior lumbar interbody fusion may have compromised their results. In our present trial, very well-matched treatment groups of a select subset of patients coming to surgery for isthmic spondylolisthesis were studied. This group with unstable slips represented less than one-third of isthmic patients operated upon during the study period and eliminated many

Table 4 Pain and functional outcome data 6

12

24

Months follow-up

AP

PL

AP

PL

AP

PL

Oswestry Disability Index (mean6SD)

22.469.3

34.3611.0

16.767.2

24.1610.2

14.669.6

18.366.2

p!.0001 Pain intensity (VAS) (mean6SD)

2.961.7

p!.0001 3.661.7

2.661.7

p5.05 AAOS (mean6SD)

79611.1

3.161.2

2.261.6

p5.06 6269.0

p5.01

p5.23

80614.4

73611.0 p5.02

2.461.1 p5.89

84613.0

8067.8 p5.56

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone; SD5standard deviation; VAS5visual analog scale (0–10); AAOS5American Academy of Orthopaedic Surgeons Modems Lumbar Questionnaire.

612

J. Swan et al. / The Spine Journal 6 (2006) 606–614

Table 5 Work and medication outcome data 6 Months follow-up Medication usage None Occasional NSAIDs Daily NSAIDs Any narcotics Work status Retired Disabled Part-time Full-time (with restrictions) Full-time (no restrictions)

12

24

AP

PL

AP

PL

AP

PL

6 11 21 9

4 2 22 18

15 11 12 9

10 7 16 13

17 15 8 7

16 13 6 11

2 12 4 4 25

1 22 8 2 13

2 7 3 3 32

1 12 6 3 24

3 3 3 7 31

3 7 6 3 72

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone; NSAID5nonsteroidal anti-inflammatory drugs.

confounding variables (eg, multisegment disc degeneration, nonmobile segment, serious neurological deficits). Furthermore, although isthmic spondylolisthesis occurs much more commonly at L5–S1, these unstable lytic segments were more likely to be at L4–L5 or at L5–S1 but above a low intercrestal line. The anterior iliolumbar ligament complex is usually absent or at a mechanical disadvantage in these rostral segments. The anterior slip, therefore, may more commonly be relatively unconstrained or unstable in these segments. A preponderance of slips requiring surgery at higher segments has been previously reported [30]. Therefore, this is a study of somewhat atypical but homogenous subjects with a common clinical problem. As opposed to previous series with heterogenous patient groups, our study found a large clinical effect of adding anterior interbody fusion to the commonly performed posterior instrumented fusion. The achievement of a highgrade functional outcome in the first year after surgery was 2–3 times more likely after the combined surgery.

All dimensions of clinical outcome, pain, disability, work resumption, and medication intake were better in the combined group through 2 years after surgery, though between-group differences attenuated appreciably after 6 months. The achievement of a subject’s minimally acceptable outcome was also better in the combined group at each follow-up point, with again notable attenuation between 6 months and 2 years. Both groups achieved some reduction of the slip angle and listhesis with restoration of some disc height. Without an anterior reconstruction, however, this correction was not maintained. As reported in other studies, the preservation of disc height and lordosis was clearly better in the combined surgery group whereas the posterior-only group returned to near preoperative status [2,11,14,28]. The progressive loss of immediate reduction over time may account, in part, for increased symptoms in the posterior fusion group. There must be a rotation and shear moment about the pedicle screws with settling of disc height and angulation of the cephalad vertebrae. It is reasonable to assume that this motion through the fixation construct may be locally painful or irritating to the exiting root. Maintenance of a more anatomic orientation has theoretical advantages for long-term outcomes by achieving better spinal balance and decreased adjacent segment stresses. Whether these supposed benefits will be realized remains to be seen with long-term studies. This study has certain limitations. As described, the study population is relatively narrow, and whether these results could be generalized to the more heterogeneous population cannot be assumed. In fact, the failure of other studies on unselected patients to find a clear clinical benefit from added anterior column support may be because the more stable slips are not helped with this maneuver. If no benefit is gained in a sizable part of a study cohort, this effect may dilute an apparent advantage occurring in the more unstable minority of subjects. Another limitation

Table 6 Primary and secondary outcomes Months follow-up

6

Treatment n

AP 47

PL 46

AP 47

PL 46

AP 47

PL 46

Achieved ODI#20

30 RR52.67 (1.53, 4.67) p5.0001 29 RR51.88 (1.10, 3.21) p5.01 29

11

34 RR51.66 (1.14, 2.42) p!.005 35 RR51.30 (0.93, 1.83) p5.12 36

20

36 RR51.21 (0.93, 1.59) p5.47 38 RR51.12 (0.82, 1.54) p5.47 42

29

Return to full-time work

Achieve minimumacceptable outcome

RR52.18 (1.31, 3.65) p5.0011

12

15

13

RR51.85 (1.27, 2.71) p5.0005

24

27

19

30

32

RR51.28 (1.04, 1.59) p5.0179

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone; ODI5Oswestry Disability Index; RR5risk ratio of AP to PL outcome (95% confidence interval), chi-square p value.

J. Swan et al. / The Spine Journal 6 (2006) 606–614

involves study design. Although the groups are wellmatched, the treatments were not randomly assigned. Sequential cohort studies may produce outcome biases having to do with temporal changes in clinical or social expectations beyond the controlled parameters (eg, improved surgeon experience or technique, economic forces effecting work availability). Furthermore, the observational (ie, nonrandomized) design precludes control of unknown prognostic factors. Despite apparent comparability, undetected differences between groups could have confounded the estimates of effect. However, such differences would have to be relatively large to completely explain the observed results, and we have no evidence suggesting that the groups are prognostically dissimilar. Although there are limitations, the study has many strengths, including the well-matched treatment groups, use of validated outcome measures, low nonresponse rate, and long-term follow-up. We feel the large observed effect is valid and will likely be reproducible in other centers in this clinical situation. The benefits of more rapid pain relief, return of function, and occupational engagement are important to the individual, their families, and society. The increase in operative time, the second incision, and increase in some complications (albeit minor) must be weighed against the early clinical benefits. Although we did not perform a formal cost assessment of the two procedures, the additional costs of increased surgical time, increased implant expenses, and treatment of minor

Table 7 Complications and reoperations

Minor complications Superficial abrasion on chin Transient ulnar nerve paresthesia without palsy Prolonged ileus (O48 hours) Atrial flutter Urinary tract infection Transient anterior thigh pain Kefzol reaction Iliac vein laceration without transfusion Major complications Deep vein thrombosis Iliac vein laceration with transfusion Transient retrograde ejaculation Superficial infection Deep infection Postoperative new-onset clinical depression Transient L5 paresis Urinary tract infection with bacteremia Non-union Reoperation Hardware removal Revision fusion Laminectomy/screw exploration

AP

PL

1 0

0 1

2 1 2 3 1 2

0 0 1 0 0 0

1 1 1 0 0 1

0 0 0 1 0 2

1 1 1

1 1 3

1 0 1

0 1 1

AP5combined anterior and posterior fusion; PL5posterior-lateral fusion alone.

613

complications are of concern. Against these procedure-related costs is the prolonged occupation disability reported in the posterior-lateral fusion group along with greater medication consumption over 2 years and the more frequent nonunions.

Conclusion Outcomes up to 2 years were superior by clinically important differences after a combined anterior-posterior operation compared with posterior-alone surgery for unstable spondylolisthesis; however, between-group differences attenuated appreciably after 6 months. The apparent clinical and occupational benefits of combined fusion should be considered along with possible increases in minor complications and procedure-related costs. References [1] Osterman K, Schlenzka D, Poussa M, Seitsalo S, Virta L. Isthmic spondylolisthesis in symptomatic and asymptomatic subjects, epidemiology, and natural history with special reference to disk abnormality and mode of treatment. Clin Orthop Relat Res 1993;297:65–70. [2] Kwon BK, Hilibrand AS, Malloy K, et al. A critical analysis of the literature regarding surgical approach and outcome for adult lowgrade isthmic spondylolisthesis. J Spinal Disord Tech 2005;18(Suppl.): S30–40. [3] Carragee EJ. Single-level posterolateral arthrodesis, with or without posterior decompression, for the treatment of isthmic spondylolisthesis in adults: a prospective, randomized study. J Bone Joint Surg Am 1997;79:1175–80. [4] de Loubresse CG, Bon T, Deburge A, Lassale B, Benoit M. Posterolateral fusion for radicular pain in isthmic spondylolisthesis. Clin Orthop Relat Res 1996;323:194–201. [5] McAfee PC, DeVine JG, Chaput CD, et al. The indications for interbody fusion cages in the treatment of spondylolisthesis: analysis of 120 cases. Spine 2005;30:S60–5. [6] Deguchi M, Rapoff AJ, Zdeblick TA. Posterolateral fusion for isthmic spondylolisthesis in adults: analysis of fusion rate and clinical results. J Spinal Disord 1998;11:459–64. [7] Hashimoto T, Shigenobu K, Kanayama M, et al. Clinical results of single-level posterior lumbar interbody fusion using the Brantigan I/F carbon cage filled with a mixture of local morselized bone and bioactive ceramic granules. Spine 2002;27:258–62. [8] Heary RF, Bono CM. Circumferential fusion for spondylolisthesis in the lumbar spine. Neurosurg Focus 2002;13. E3. [9] Kwon BK, Albert TJ. Adult low-grade acquired spondylolytic spondylolisthesis: evaluation and management. Spine 2005;30:S35–41. [10] La Rosa G, Conti A, Cacciola F, et al. Pedicle screw fixation for isthmic spondylolisthesis: does posterior lumbar interbody fusion improve outcome over posterolateral fusion? J Neurosurg 2003;99: 143–50. [11] Madan S, Boeree NR. Outcome of posterior lumbar interbody fusion versus posterolateral fusion for spondylolytic spondylolisthesis. Spine 2002;27:1536–42. [12] Moller H, Hedlund R. Instrumented and noninstrumented posterolateral fusion in adult spondylolisthesisda prospective randomized study: part 2. Spine 2000;25:1716–21. [13] Potter BK, Freedman BA, Verwiebe EG, Hall JM, Polly DW Jr, Kuklo TR. Transforaminal lumbar interbody fusion: clinical and radiographic results and complications in 100 consecutive patients. J Spinal Disord Tech 2005;18:337–46.

614

J. Swan et al. / The Spine Journal 6 (2006) 606–614

[14] Suk KS, Jeon CH, Park MS, Moon SH, Kim NH, Lee HM. Comparison between posterolateral fusion with pedicle screw fixation and anterior interbody fusion with pedicle screw fixation in adult spondylolytic spondylolisthesis. Yonsei Med J 2001;42:316–23. [15] Axelsson P, Johnsson R, Stromquist B. Is there increased intervertebral motility in isthmic adult spondylolisthesis? Spine 2000;25: 1701–3. [16] McGregor AH, Anderton L, Gedroyc WM, Johnson J, Hughes SP. The use of interventional open MRI to assess the kinematics of the lumbar spine in patients with spondylolisthesis. Spine 2002;27: 1582–6. [17] Floman Y. Progression of lumbosacral isthmic spondylolisthesis in adults. Spine 2000;25:342–7. [18] Schneider G, Pearcy M, Bogduk N. Letters. Spine 2005;30:2474–5. [19] Schneider G, Pearcy MJ, Bogduk N. Abnormal motion in spondylolytic spondylolisthesis. Spine 2005;30:1159–64. [20] D’Andrea G, Ferrante L, Dinia L, Caroli E, Orlando ER. ‘‘Supineprone’’ dynamic X-ray examination: new method to evaluate lowgrade lumbar spondylolisthesis. J Spinal Disord Tech 2005;18:80–3. [21] Lumbar Spine Instrument. Data Outcomes Collection Package. version 1.4 ed. Chicago, IL: American Academy of Orthopaedic Surgeons Council of Musculoskeletal Specialty Societies Council of Spine Societies, 1996. [22] Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66:271–3. [23] Main C, Wood P, Hollis S, Spanswick C, Waddell G. The distress and risk assessment method (DRAM): a simple patient classification to

[24]

[25]

[26]

[27]

[28]

[29]

[30]

identify distress and evaluate the risk of a poor outcome. Spine 1992;17:42–52. Boos N, Semmer N, Elfering A, et al. Natural history of individuals with asymptomatic disc abnormalities in magnetic resonance imaging: predictors of low back pain–related medical consultation and work incapacity. Spine 2000;25:1484–92. Jensen M, Brant-Zawadzki M, Obuchowski N, Modic M, Malkasian D, Ross J. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med 1994;331:69–73. Carragee EJ, Tanner CM, Khurana S, et al. The rates of false-positive lumbar discography in select patients without low back symptoms. Spine 2000;25:1373–80; discussion 1381. Weishaupt D, Zanetti M, Hodler J, et al. Painful lumbar disk derangement: relevance of endplate abnormalities at MR imaging. Radiology 2001;218:420–7. Kwon BK, Berta S, Daffner SD, et al. Radiographic analysis of transforaminal lumbar interbody fusion for the treatment of adult isthmic spondylolisthesis. J Spinal Disord Tech 2003;16:469–76. Carragee E, Alamin TF. A prospective assessment of patient expectations and satisfaction in spinal fusion surgery. Proceedings International Society for the Study of the Lumbar Spine 2003; Vancouver, BC: 74. Spruit M, Pavlov PW, Leitao J, De Kleuver M, Anderson PG, Den Boer F. Posterior reduction and anterior lumbar interbody fusion in symptomatic low-grade adult isthmic spondylolisthesis: short-term radiological and functional outcome. Eur Spine J 2002; 11:428–33.