- Email: [email protected]
The effect of a radiographic solid fusion on clinical outcomes after minimally invasive transforaminal lumbar interbody fusion
The Spine Journal 11 (2011) 205–212
Clinical Study
The effect of a radiographic solid fusion on clinical outcomes after minimally invasive transforaminal lumbar interbody fusion Yung Park, MD*, Joong Won Ha, MD, Yun Tae Lee, MD, Na Young Sung, MS Department of Orthopedic Surgery, Yonsei University College of Medicine, National Health Insurance Medical Center, 1232, Baeksok St, Ilsan district, Goyang city, Gyeonggi province, 410-719, Republic of Korea Received 20 May 2010; revised 21 December 2010; accepted 26 January 2011
Abstract
BACKGROUND CONTEXT: The correlation between radiographic solid fusion and favorable clinical outcome has not been fully established. Many surgeons believe that patients who achieve a radiographic solid fusion will exhibit a more positive clinical outcome than those getting an unsuccessful fusion. To our knowledge, there is no study that has evaluated whether a solid fusion influences clinical outcome after minimally invasive lumbar fusion. PURPOSE: This study was designed to evaluate the effect of radiographic solid fusion on clinical outcome after minimally invasive transforaminal lumbar interbody fusion (TLIF). STUDY DESIGN: We conducted a retrospective study by comparing the prospectively collecting data. PATIENT SAMPLE: The sample comprises 66 patients who had achieved a solid fusion or nonunion at least 2 years after minimally invasive TLIF for the treatment of low-grade spondylolisthesis or degenerative segmental instability. OUTCOME MEASURES: The outcome measures were visual analog scale (VAS) for back pain and radiating leg pain, Oswestry Disability Index (ODI), functional scale (defined as a modified method of Whitecloud et al.), and radiographic fusion status. METHODS: Two independent spine surgeons reviewed the completed medical records and radiographic data of 66 patients who had undergone minimally invasive TLIF by one surgeon at an institution. Clinical outcome was evaluated using VAS, ODI, and functional scale. The radiographic fusion status was assessed using flexion-extension lateral radiographs and computed tomography scans. Comparison and correlation analyses were performed to examine the relationship between fusion status and clinical outcome. RESULTS: There were 51 (77%) patients in the solid fusion (control) group and 15 patients in the nonunion group. The improvement from baseline with regard to VAS scores for back and leg pain as well as ODI scores was significant in both groups (all, p!.0001), with patients in the control group reporting significantly better improvement of back pain scores than those in nonunion group (p5.04). Conversely, the improvement of VAS scores for leg pain and ODI scores was comparable between two groups. Forty-one patients (80%) in the control group and 13 (87%) in the nonunion group demonstrated an excellent or good result in the final functional scale as well. The presence of radiographic solid fusion positively, but not strongly, correlated with the improvement of VAS scores for back pain (r50.255, p5.039). CONCLUSIONS: At least 2 years after minimally invasive TLIF, better reduction of back pain was noted in patients who achieved a radiographic solid fusion as opposed to those with nonunion. However, there was no clear evidence that radiographic solid fusion was associated with better clinical outcome scores or improvement in leg pain than nonunion. Crown Copyright Ó 2011 Published by Elsevier Inc. All rights reserved.
Keywords:
Radiographic solid fusion; Clinical outcome; Minimally invasive; TLIF
FDA device/drug status: not applicable. Author disclosures: none. * Corresponding author. Department of Orthopedic Surgery, Yonsei University College of Medicine, National Health Insurance Medical Center,
1232, Baeksok St, Ilsan district, Goyang city, Gyeonggi province, 410719, Republic of Korea. Tel.: (82) 31-900-0270; fax: (82) 31-900-0343. E-mail address: [email protected] (Y. Park)
1529-9430/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2011.01.023
206
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Context Several studies over the years have aimed to determine if the attainment of a demonstrably solid fusion correlates with clinical outcomes. Given the methodological limitations of many of these studies, the question remains open. In this article, the authors aim to gain some insight on this issue in patients undergoing transforaminal lumbar interbody fusion. Contribution The authors found that at midterm follow-up, patients who attained a solid fusion by their definition demonstrated improvements in leg pain and functional outcome similar to those who failed to fuse. However, the fusion group had greater improvement in back pain on VAS measurement. Paradoxically, the classification of ‘‘good-to-excellent results’’ was slightly better in the ‘‘nonunion’’ group. Implication The authors clearly outlined the limitations of the study in their discussion. The question addressed remains important in light of the fact that increasingly aggressive efforts are being made at achieving solid fusion, despite the fact that many randomized controlled trials (including SPORT) now suggest that more aggressive fusions do not necessarily improve outcomes as a whole for patients with degenerative indications. —The Editors
Introduction Minimally invasive lumbar fusion surgery has been reported to potentially offer a number of benefits over traditional open procedure, including less blood loss, less softtissue trauma, less postoperative back pain, shorter hospital stay, and earlier return to work [1–9]. In addition to these purported advantages, the achievement of a solid fusion is considered to be an ultimate surgical goal of minimally invasive spine fusion surgery as well as with open surgery. Many surgeons believe that patients achieving a radiographic solid fusion will exhibit more positive clinical outcome than those getting an unsuccessful radiographic fusion. Unlike other studies [10,11] that indicated no relation between radiographic solid fusion and clinical outcome, Kornblum et al. [12] found that successful fusion correlated with improved long-term clinical results in regard to back pain and lower limb function. To our knowledge, however, there is no study that has evaluated whether a radiographic solid fusion influences the clinical outcome of minimally invasive lumbar spinal
fusion. Our hypothesis is that a favorable clinical outcome is attained more often in patients achieving radiographic solid fusion rather than those failing to achieve radiographic fusion after minimally invasive spine fusion surgery. The purpose of this study was to evaluate the effect of radiographic solid fusion through minimally invasive transforaminal lumbar interbody fusion (TLIF) procedure on clinical outcome.
Materials and methods Patients’ population We performed a retrospective study by comparing the prospectively collecting data from consecutive patients who had achieved a solid fusion versus nonunion at least 2 years after minimally invasive TLIF by a single surgeon (YP) at an academic institution from 2004 to 2006. The surgical indications were patients who were suffering from low back pain and radiating pain down to lower extremity (leg pain) associated with a single-level lumbar or lumbosacral segmental instability: adult acquired lowgrade (Grade I/II) spondylolytic spondylolisthesis; Grade I/II degenerative spondylolisthesis with gross instability; and degenerative segmental instability combined with lumbar stenosis (central, lateral, and foraminal stenosis) and/or lumbar disc herniation. The inclusion criteria of segmental instability was $4 mm of translation or $10 of angular motion on preoperative flexion and extension radiographs. Our relative contraindications for minimally invasive approach, which were treated with traditional open surgery, included the following: high-grade (Grade III/IV) spondylolisthesis; patients with severely collapsed disc space and no motion on flexion-extension radiographs; patients who needed a multilevel decompression and fusion; those who had combined fixed coronal and/or sagittal deformities (kyphoscoliosis) that needed a surgical correction; or those who had a disease involving trauma, infection, and pathologic causes. All patients had preoperative evaluation with static and dynamic plain lumbar radiographs, magnetic resonance imaging, and/or computed tomography (CT). The surgery was undertaken when the preoperative symptoms and signs of back and leg pain were refractory to nonoperative treatment, such as medications, physiotherapy, epidural steroid injection, and/or a neurologic deterioration was developed. The surgical goal of all patients was a decompression and fusion for the treatment of back and leg pain arising from lumbar segmental instability. Surgical technique Under fluoroscopic guidance, an appropriate length of 22-mm–diameter METRx (Medtronic, Memphis, TN, USA) tubular retractor was introduced through a 2.5-cm incision for both neural decompression and access to the
Y. Park et al. / The Spine Journal 11 (2011) 205–212
interbody space. The approach was carried out on the side of the worst preoperative symptom of radiculopathy. Sextant screws and rod were placed for percutaneous pedicle screw fixation. A more detailed description of the procedure is available in the literatures [1–9]. In all cases, the autogenous bone obtained from the resected lamina and facet mixed with demineralized bone matrix (OsteofilRT DBM paste; Regeneration Technologies Inc., Alachua, FL, USA) was placed anteriorly and contralateral to the annulotomy within the interbody space, and then a polyetheretherketone cage (Capstone; Medtronic, Memphis, TN, USA) was inserted into the disc space. No additional contralateral facet fusion was performed in all patients. Data assessment Clinical outcome was evaluated using the patient-assessed quantitative measurement of visual analog scale (VAS) for back/leg pain down to lower extremity and Oswestry Disability Index (ODI), as well as the surgeon-assessed outcome measurement of a modified functional scale [13] (Table 1), both preoperatively and at each postoperative follow-up. The VAS scores were recorded on a 10-mm horizontal linewith 0 equal to ‘‘no pain’’ and 10 equal to ‘‘very severe pain.’’ The ODI was scored on a 0 to 100 scale using the Oswestry Disability Questionnaire. The subjective postoperative symptoms documented at each postoperative visit were divided into four categories: symptom free, back pain only, leg pain only, and both back and leg pain. Standing anteroposterior, lateral, flexion, and extension radiographs of the lumbosacral spine were collected from preoperative and final postoperative visit for the fusion assessment. Postoperative CT scan was also obtained at the final visit. All the clinical and radiographic data were reviewed by independent experienced spine surgeons (JWH, YTL) for analyses. A radiographic solid fusion was determined by following criteria that were proposed by Burkus et al. [14]; no motion (acceptable intraobserver measurement error was 3 angular motion or 3 mm of translation) on flexionextension lateral radiographs; continuous bony bridge within/around the cage (incorporation of the grafted bone into the vertebral end plates) on the CT scan (Figure, Top
207
Left, Top Right); lack of radiolucent lines around the graft and cage as well as absence of a lucent hollow around the pedicle screws on dynamic radiographs and/or CT scan; and new bone formation adjacent to or within the cage and/or fused posterior facet joint (the opposite side of TLIF approach) on the CT scan (Figure, Top Right, Bottom). The fusion status was subdivided into four grades: Grade 1 (definitely solid)—no motion on flexion-extension radiographs, continuous bony bridge within/around the cage, new bone formation adjacent to or within the cage, and/or fused posterior facet joint on CT scan; Grade 2 (possibly solid)—no motion on dynamic radiographs—and continuous bony incorporation within/around the cage, without evidence of new bone formation adjacent to or within the cage or facet joint fusion; Grade 3 (probably not solid)—no motion excluding evidence of bony incorporation within/around the cage; and Grade 4 (definitely not solid)—motion on dynamic radiographs with no evidence of bony bridge within/around the cage. According to our fusion grades, we considered the Grades 1 and 2 as a radiographic solid fusion and the Grades 3 and 4 as a nonunion (Table 2). Statistical methods Patients were divided into two groups: a radiographic solid fusion (control) group (Grades 1 and 2) and nonunion group (Grades 3 and 4). Comparison and correlation analyses were performed to evaluate the relationship between fusion status and clinical outcome. The demographics and characteristics and validated outcome measurements (such as VAS, ODI score, and functional scale) were compared between two groups. The comparisons between the groups were based on chi-square tests or Fisher exact tests for categorical variables, and on independent t tests for continuous variables. Paired t tests were used to evaluate the clinical improvement after surgery by analyses of changes over time of VAS and ODI scores. The phi coefficient was used for a measure of the association between two categorical variables, and the point-biserial correlation was also used to determine the correlation between a continuous variable and a categorical variable. In all analyses, a probability less than .05 considered to be statistically significant. The analyses were performed with the use of SPSS version 10.0 (SPSS Inc., Chicago, IL, USA).
Table 1 Criteria for the assessment of functional outcome Outcome
Pain
Medication
Activity
Work status
Excellent
None except for occasional back pain Markedly improved, occasional pain Some improvement No change in symptoms or a worsening of the patient’s condition
None
Normal
Normal
Occasional use of pain medication
Minimal functional limitations
Frequent use of pain medication Oral use of narcotics
Restricted Incapacitated
Return to work, although not at the same job activity Limited Disabled
Good Fair Poor
208
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Figure. A radiographic solid fusion demonstrated on sagittal and coronal images of computed tomography scan: (Top Left) continuous bony incorporation within and around cage; (Top Right) new bone formation at the posterior margin of interbody space; and (Bottom) fused posterior facet joint.
Results A total of 76 patients comprised the inclusion criteria of our study. Two patients died during the follow-up period without causes related to index surgery, and eight declined Table 2 Grades for the assessment of fusion status CT finding
Grade
No motion on flexionextension radiographs
Continuous bony incorporation within and/or around cage
New bone formation adjacent to or within the cage and/or fused posterior facet joint
Definitely solid Possibly solid Probably not solid Definitely not solid
Yes Yes Yes No
Yes Yes No No
Yes No No No
CT, computed tomography.
to participate the radiographic and CT evaluation at final follow-up. Finally, 66 patients (87%) completed the follow-up visit and were included in the analysis. There were 20 men and 46 (69.7%) women with an average age of 57.569.2 years (range, 40–81 years). The mean duration of follow-up was 36.169.9 months (range, 24–63 months). Thirteen patients (19.7%) were current smokers. Coexisting conditions included 24 (36.4%) hypertension, 10 (15.2%) diabetes, and 16 (24.2%) osteoporosis (t score #2.5). The average duration of preoperative symptoms was 2 years (range, 0.5–20 years). There were 24 degenerative spondylolisthesis, 23 spondylolytic spondylolisthesis, and 19 degenerative lumbar instabilities. The most common segment treated was L4–L5 (40/66, 61%), followed by L5–S1 (21/66, 32%), and then L3–L4 (5/66, 7%). There were no conversions to an open procedure. According to our criteria, the fusion status was demonstrated as following four grades: Grade 1 (definitely solid)
Y. Park et al. / The Spine Journal 11 (2011) 205–212
209
Table 3 Preoperative clinical features of nonunion group and controls who had a solid fusion Demographics and characteristics
Overall
Controls
Number of patients
66
51
15
Mean age (range) (y)
57.5 (40–81)
56.8 (40–81)
60.3 (45–74)
.19
Female gender (N)
46
36
10
.76y
Mean body mass index (range) (kg/m2) #25 (N) O25 (N)
24.6 (18.2–34.6) 37 29
24.4 (18.2–34.6) 29 22
25.2 (21.1–30.9) 8 7
.38 NSy
Mean height (range) (cm)
156.6 (142–178)
156 (142–176)
159.7 (145–178)
.15
ASA class 1/2/3/4 (N)
23/43/0/0
19/32/0/0
4/11/0/0
.65y
Patients who were smokers (N)
13
10
3
NSy
Patients who had osteoporosis (T#2.5) (N)
16
13
3
NSy
Mean duration of symptoms (range) (y)
2 (0.5–20)
1.8 (0.5–10)
2.8 (0.5–20)
.21
Mean periods of follow-up (range) (mo)
36.1 (24–63)
36.7 (24–63)
33.7 (24–52)
.31
Preoperative diagnosis (N) Spondylolytic spondylolisthesis (Meyerding Grade 1, Grade 2) Degenerative spondylolisthesis (Meyerding Grade 1, Grade 2) Degenerative lumbar instability Level of surgery (N) L3–L4 L4–L5 L5–S1
Nonunion
p Value*
.41y 23 (12, 11) 24 (22, 2) 19
17 (7, 10) 17 (15, 2) 17
6 (5, 1) 7 (7, 0) 2 .6y
5 40 21
3 32 16
2 8 5
N, number of patients; NS, not significant; ASA, American society of anesthesiologists. * The p values are based on chi-square or Fisher exact tests for categorical variables and on independent t tests for continuous variables. y The distribution of variables was not significantly different.
in 33 patients (50%); Grade 2 (possibly solid) in 18 (27.3%); Grade 3 (probably not solid) in 2 (3%); and Grade 4 (definitely not solid) in 13 (19.7%). On flexion-extension radiographs, no motion was demonstrated in 53 (80.3%) patients. A lucent hollow around the pedicle screws was found in 11 (84.6%) of 13 patients of Grade 4 and none of Grade 3. The final radiographic fusion rate of our series was 77% (51 patients of Grades 1 and 2). Nonunion was allocated to 15 patients (23%) of Grades 3 and 4. None of them undertook revision surgery to repair nonunion. Fifty-one patients comprised a solid fusion (control) group, and 15 patients were included in the nonunion group. The preoperative factors, such as age, gender, body mass index, general health status classified by American Society of Anesthesiologists, smoking, osteoporosis, preoperative diagnosis, and level of surgery, were similar between two groups. The detailed preoperative demographics and characteristics were illustrated in Table 3. The average pre- and postoperative final VAS (for back and leg pain) and ODI scores were comparable between the groups. Each preoperative VAS for back and leg pain and ODI scores were significantly improved at final postoperative follow-up in both groups (all measurements, p!.0001) (Table 4).
Forty-one patients (80%) in the control group and 13 (87%) in the nonunion group demonstrated an excellent or good result of the final functional scale [14]. The rate of solid fusion was 76% (41/54) in patients with excellent or good result and 83% (10/12) in patients with fair or poor result (statistically not significant) (Table 4). Twenty-five (49%) patients in the control group and eight (53%) in the nonunion group revealed none of the symptoms at final visit. Both groups revealed similar distributions of final functional outcome (excellent and good vs. fair and poor) and postoperative symptoms (symptom free or not) (Table 4). Nevertheless, the postoperative improvement of VAS for back pain was significantly higher (p5.04) in the control group (3.9, 95% confidence interval of the paired difference, 3.2–4.6) than in the nonunion group (2.5, 95% confidence interval of the paired difference, 1.6–3.4). On the other hand, the comparable improvement of VAS for leg pain and ODI score is demonstrated between the two groups (Table 4). The presence of radiographic solid fusion positively, but not strongly, correlated with a postoperative improvement (paired difference) of VAS for back pain (r50.255, p5.039). On the contrary, it did not correlate with the
210
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Table 4 Clinical outcome of nonunion group and controls who had a solid fusion Fusion status Clinical outcome
Overall
Controls
Nonunion
Number of patients
66
51
15
VAS for back pain (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
6.261.9 (3–10) 2.662.1 (0–8) 3.662.4 3.0–4.2 !.0001
6.462.0 (3–10) 2.562.1 (0–8) 3.962.5 3.2–4.6 !.0001
5.361.6 (3–8) 2.962.3 (0–7) 2.561.6 1.6–3.4 !.0001
.06 .58 .04
VAS for leg pain (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
8.162.0 (3–10) 1.662.8 (0–9) 6.563.1 5.8–7.3 !.0001
8.162.1 (3–10) 1.662.9 (0–9) 6.563.2 5.9–7.4 !.0001
8.361.6 (5–10) 1.762.7 (0–7) 6.762.8 5.1–8.2 !.0001
.67 .93 .85
ODI score (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
60.2616.5 (24–88) 25.9617.9 (0–78) 34.3620.2 29.3–39.2 !.0001
60.7617.5 (24–88) 26.6619.1 (0–78) 34.1621.9 28.0–40.3 !.0001
58.3613.0 (40–82) 23.6613.3 (6–56) 34.7613.4 27.3–42.1 !.0001
.62 .58 .93
Final functional scale, N (%) Excellent Good Fair Poor
37 17 10 2
(56) (26) (15) (3)
28 13 8 2
(55) (25) (16) (4)
9 (60) 4 (27) 2 (13) 0
.86x,k
Final postoperative symptoms, N (%) Free of symptoms Back pain Leg pain Back and leg pain
33 16 4 13
(50) (24) (6) (20)
25 14 3 9
(49) (27) (6) (18)
8 2 1 4
.69x,k
(53) (13) (7) (27)
p Value*
VAS, visual analog scale; ODI, Oswestry Disability Index; CI, confidence interval. * The p values are based on independent t tests for continuous variables. y The values are given as the mean and the standard deviation. z The p values are based on paired t tests for the analyses of changes over time. x The p values are based on chi-square tests for categorical variables (excellent+good vs. fair+poor, free of symptoms vs. symptomatic). k The distribution of variables was not significantly different.
other variables of clinical outcome, such as final VAS and ODI score, postoperative symptoms, and functional scale (Table 5). The group with degenerative lumbar instabilities showed a higher fusion rate (89.5%) than another two groups with spondylolisthesis (73.9% in spondylolytic spondylolisthesis group and 70.8% in degenerative spondylolisthesis group, respectively). The difference, however, was not statistically significant (p5.35) in regard to fusion status among three different preoperative diagnoses (Table 6).
Discussion Minimally invasive TLIF procedure allows surgeon to preserve posterior spinal–stabilizing structures including lamina, unilateral facet joint, paraspinal muscles, and posterior ligaments [1–9]. For the reason that preserving these
structures can be helpful to sustain spinal stability even in the condition of pseudarthrosis, patients with radiographicfailed fusion after minimally invasive procedure may have not so much of a complaint in pain and disability as those having nonunion after open fusion surgery. Although the influence between fusion status and clinical outcome after traditional open posterior lumbar fusion has been examined extensively through clinical and radiographic studies [10–12,15–19], there are no series that have reported this topic after minimally invasive lumbar fusion, using multiple validated outcome instruments and multimodal radiographic assessment including high resolution CT scan. To assess the relationship of radiographic fusion with the clinical result, independent experienced spine surgeons examined a consecutive series of patients who had had minimally invasive TLIF performed by a single surgeon. In the present study, radiographic solid fusion was achieved in 77% of patients, and overall functional
Y. Park et al. / The Spine Journal 11 (2011) 205–212 Table 5 Correlations between fusion status and clinical outcome Fusion status
Clinical outcome
Correlations
The presence of solid fusion or not
VAS for final back pain
Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Phi coefficient p Value N Phi coefficient p Value N
0.07 .576 66 0.255* .039 66 0.012 .926 66 0.024 .848 66 0.07 .576 66 0.011 .93 66 0.15 .743 66 0.069 .925 66
Paired difference of VAS for back pain VAS for final leg pain
Paired difference of VAS for leg pain Final ODI score
Paired difference of ODI score Final postoperative symptoms Final modified functional scale by Whitecloud et al. [13]
N, number of valid cases; VAS, visual analog scale; ODI, Oswestry Disability Index. * Correlation is significant at the 0.05 level (two tailed).
improvement was noted in 82% of patients after minimally invasive TLIF. Patients in both groups exhibited highly significant improvements in back pain and radiating pain down to lower extremity (leg pain) and disability compared with their preoperative baseline. Additionally, the present study demonstrates that patients who achieved solid fusion appear to have a greater improvement of back pain from baseline than those with nonunion, although it did not show a strong correlation. On the contrary, the improvement of leg pain and disability and functional outcome seem to be independent from the radiographic fusion. Table 6 Radiographic fusion status among three subgroups after minimally invasive TLIF Overall Number of patients 66
Group 1
Group 2
Group 3
23
24
19
p Value*
Fusion status Solid fusion, N (%) 51 (77.3) 17 (73.9) 15 (70.8) 17 (89.5) .35 Nonunion, N (%) 15 (22.7) 6 (26.1) 7 (29.2) 2 (10.5) TLIF, transforaminal lumbar interbody fusion. The values are given as the number of patients. Group 1: spondylolytic spondylolisthesis. Group 2: degenerative spondylolisthesis. Group 3: degenerative lumbar instability. * The p values are based on chi-square tests for the analyses of categorical variables.
211
Burkus et al. [14] suggested that new bone formation adjacent to or within the cage is the most reliable radiographic indication of a solid fusion. Additionally, they also proposed the other criteria substantiating a radiographic solid fusion: no significant motion at the instrumented spinal segment on dynamic studies; incorporation of grafted bone into the vertebral end plates; and absence of progressive radiographic lucent lines around the cage. Our inclusion criteria for a radiographic solid fusion contained these reliable radiographic clues. Although our fusion Grade 2 is literally ‘‘possibly solid,’’ and this grade does not include the criterion of new bone formation around the interbody space or facet joint fusion, this category obviously meets the other generally accepted criteria to substantiate a radiographic solid fusion as it mentioned as the above: no motion on flexion-extension lateral radiographs; continuous bony bridge within/around the cage on the CT scan; lack of radiolucent lines around the graft and cage; and absence of lucent hollow around the pedicle screws [14]. For that reason, we believe that our Grades 1 and 2 are considered as a radiographic solid fusion. On the contrary, fusion Grade 3 does not demonstrate any evidence of continuous bony incorporation within or around the cage but just grafted bone observed within the cage on the CT scan. Although the Grade 3 has no motion on dynamic radiographs at the instrumented spinal motion segment, this category corroborates nonunion. As we have used such a rigorous definition of fusion, the radiographic solid fusion was demonstrated only in 77% of our series. In the previous studies [2,5,7,9], almost 100% of fusion rate was reported based on the flexion-extension radiographs and/or postoperative CT scan at 2 years after minimally TLIF and significantly improved outcomes were also noted. In particular, recombinant human bone morphogenetic protein-2 soaked in a collagen sponge carrier was used as a graft material in these series. Direct comparison of our fusion rate with others appears to be inadequate because of the different graft material and methods of measurement. Application of rigorous fusion criteria and nonuse of recombinant human bone morphogenetic protein-2 might explain the relatively low rate of fusion compared with historical studies [2,5,7,9]. In the present study, postoperative back pain, leg pain, and disability were comparable in the radiographic solid fusion versus nonunion group. The patients with radiographic nonunion did not have so much of a complaint in pain and disability, rendering revision to repair nonunion not necessary. Although the final postoperative back pain scores were similar in both groups, considering the improvement scale from the preoperative to the postoperative period, a significantly better improvement was noted in the solid fusion group, whose correlation was not so strong statistically. It might be coming from one of the reasons that reducing iatrogenic soft-tissue damage by minimally invasive lumbar fusion procedure could contribute to decrease the back pain and disability of patients even
212
Y. Park et al. / The Spine Journal 11 (2011) 205–212
whose radiographic solid fusions were not achieved. However, this was not statistically confirmed by comparison with traditional open lumbar fusion in the present study, which is going to be the future topic of our study. As with other studies, the present study has several limitations. Although the clinical and radiographic data were collected prospectively, the present study is retrospective with a heterogeneous population, and the patients were not randomly selected. A more homogenous population might have provided a stronger correlation between independent variables and radiographic fusion status. In addition, the cohort size of the study is not large enough to have a sufficient statistical power for complete conclusions in some aspects. Another potential limitation of the study is that our average period of follow-up is not sufficiently long to validate the effect of minimally invasive lumbar fusion on the outcomes such as adjacent segment degeneration. Furthermore, given the rarity of minimally invasive spinal fusion surgeries performing at multiple levels, the data from multiple segmental lumbar fusions were not available for the assessment as well. Additionally, the present study did not examine the radiographic measurements to determine postoperative sagittal alignment, lumbar lordosis, and correction of a slip and collapsed disc space. Finally, our study has no control group for comparison. A matched cohort of patients underwent traditional open surgery that would provide the most ideal control group. We would then be able to assess the difference of long-term clinical and radiographic results between minimally invasive and traditional open procedures. Despite of these weaknesses, the present study offers insight into the treatment results of minimally invasive TLIF and estimates the rate of a radiographic solid fusion at minimum 2 years after minimally invasive TLIF. Although our findings could not identify risk factors associated with subsequent development of nonunion, our data provide important information regarding midterm results of minimally invasive TLIF. Our hypothesis is disputed against the results of the present study so that a favorable clinical outcome is not always more likely attained in patients achieving a radiographic solid fusion rather than those developing failed fusion after minimally invasive spine fusion surgery. As a result of the present study, after at least 2 years after minimally invasive TLIF, better reduction of back pain was noted in patients who achieved a radiographic solid fusion as opposed to those with nonunion. However, there was no clear evidence that radiographic solid fusion was associated with better clinical outcome scores or improvement of leg pain than nonunion. References [1] Dhall SS, Wang MY, Mummaneni PV. Clinical and radiographic comparison of mini-open transforaminal lumbar interbody fusion
[2] [3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
with open transforaminal lumbar interbody fusion in 42 patients with long-term follow-up. J Neurosurg Spine 2008;9:560–5. Foley KT, Holly LT, Schwender JD. Minimally invasive lumbar fusion. Spine 2003;28(15 Suppl):S26–35. Isaacs RE, Podichetty VK, Santiago P, et al. Minimally invasive microendoscopy-assisted transforaminal lumbar interbody fusion with instrumentation. J Neurosurg Spine 2005;3:98–105. Mummaneni PV, Rodts GE Jr. The mini-open transforaminal lumbar interbody fusion. Neurosurgery 2005;57(4 Suppl):256–61. Park P, Foley KT. Minimally invasive transforaminal lumbar interbody fusion with reduction of spondylolisthesis: technique and outcomes after a minimum of 2 years’ follow-up. Neurosurg Focus 2008;25:1–7. Park Y, Ha JW. Comparison of one-level posterior lumbar interbody fusion performed with a minimally invasive approach or a traditional open approach. Spine 2007;32:537–43. Peng CWB, Yue WM, Poh SY, et al. Clinical and radiological outcomes of minimally invasive versus open transforaminal lumbar interbody fusion. Spine 2009;34:1385–9. Scheufler KM, Dohmen H, Vougioukas VI. Percutaneous transforaminal lumbar interbody fusion for the treatment of degenerative lumbar instability. Neurosurgery 2007;60(4 Suppl):203–12. Schwender J, Holly LT, Rouben DP, Foley KT. Minimally invasive transforaminal lumbar interbody fusion (TLIF). J Spinal Disord Tech 2005;18(Suppl):S1–6. Herkowitz HN, Kurz LT. Degenerative lumbar spondylolisthesis with spinal stenosis. A prospective study comparing decompression with decompression and intertransverse process arthrodesis. J Bone Joint Surg Am 1991;73:802–8. Fischgrund JS, Mackay M, Herkowitz HN, et al. 1997 Volvo Award winner in clinical studies. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective, randomized study comparing decompressive laminectomy and arthrodesis with and without spinal instrumentation. Spine 1997;22:2807–12. Kornblum MB, Fischgrund JS, Herkowitz HN, et al. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective longterm study comparing fusion and pseudarthrosis. Spine 2004;29: 726–33. Whitecloud TS, Davis JM, Olive PM. Operative treatment of the degenerated segment adjacent to a lumbar fusion. Spine 1994;19: 531–6. Burkus JK, Foley KT, Haid R, LeHuec JC. Surgical interbody research group—radiographic assessment of interbody fusion devices: fusion criteria for anterior lumbar interbody surgery. Neurosurg Focus 2001;10:E11. Christensen FB, Hansen ES, Eiskjær SP, et al. Circumferential lumbar spinal fusion with Brantigan cage versus posterolateral fusion with titanium Cotrel-Dubousset instrumentation: a prospective, randomized clinical study of 146 patients. Spine 2002;27: 2674–83. Diedrich O, Perlick L, Schmitt O, Kraft CN. Radiographic characteristics on conventional radiographs after posterior lumbar interbody fusion: comparative study between radiotranslucent and radiopaque cages. J Spinal Disord 2001;14:522–32. Greenough CG, Peterson MD, Hadlow S, Fraser RD. Instrumented posterolateral lumbar fusion. Results and comparison with anterior interbody fusion. Spine 1998;23:479–86. Wetzel FT, Brustein M, Phillips FM, Trott S. Hardware failure in an unconstrained lumbar pedicle screw system: a 2-year follow-up study. Spine 1999;24:1138–43. Resnick DK, Choudhri TF, Dailey AT, et al. Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 5: correlation between radiographic and functional outcome. J Neurosurg Spine 2005;2:658–61.
Clinical Study
The effect of a radiographic solid fusion on clinical outcomes after minimally invasive transforaminal lumbar interbody fusion Yung Park, MD*, Joong Won Ha, MD, Yun Tae Lee, MD, Na Young Sung, MS Department of Orthopedic Surgery, Yonsei University College of Medicine, National Health Insurance Medical Center, 1232, Baeksok St, Ilsan district, Goyang city, Gyeonggi province, 410-719, Republic of Korea Received 20 May 2010; revised 21 December 2010; accepted 26 January 2011
Abstract
BACKGROUND CONTEXT: The correlation between radiographic solid fusion and favorable clinical outcome has not been fully established. Many surgeons believe that patients who achieve a radiographic solid fusion will exhibit a more positive clinical outcome than those getting an unsuccessful fusion. To our knowledge, there is no study that has evaluated whether a solid fusion influences clinical outcome after minimally invasive lumbar fusion. PURPOSE: This study was designed to evaluate the effect of radiographic solid fusion on clinical outcome after minimally invasive transforaminal lumbar interbody fusion (TLIF). STUDY DESIGN: We conducted a retrospective study by comparing the prospectively collecting data. PATIENT SAMPLE: The sample comprises 66 patients who had achieved a solid fusion or nonunion at least 2 years after minimally invasive TLIF for the treatment of low-grade spondylolisthesis or degenerative segmental instability. OUTCOME MEASURES: The outcome measures were visual analog scale (VAS) for back pain and radiating leg pain, Oswestry Disability Index (ODI), functional scale (defined as a modified method of Whitecloud et al.), and radiographic fusion status. METHODS: Two independent spine surgeons reviewed the completed medical records and radiographic data of 66 patients who had undergone minimally invasive TLIF by one surgeon at an institution. Clinical outcome was evaluated using VAS, ODI, and functional scale. The radiographic fusion status was assessed using flexion-extension lateral radiographs and computed tomography scans. Comparison and correlation analyses were performed to examine the relationship between fusion status and clinical outcome. RESULTS: There were 51 (77%) patients in the solid fusion (control) group and 15 patients in the nonunion group. The improvement from baseline with regard to VAS scores for back and leg pain as well as ODI scores was significant in both groups (all, p!.0001), with patients in the control group reporting significantly better improvement of back pain scores than those in nonunion group (p5.04). Conversely, the improvement of VAS scores for leg pain and ODI scores was comparable between two groups. Forty-one patients (80%) in the control group and 13 (87%) in the nonunion group demonstrated an excellent or good result in the final functional scale as well. The presence of radiographic solid fusion positively, but not strongly, correlated with the improvement of VAS scores for back pain (r50.255, p5.039). CONCLUSIONS: At least 2 years after minimally invasive TLIF, better reduction of back pain was noted in patients who achieved a radiographic solid fusion as opposed to those with nonunion. However, there was no clear evidence that radiographic solid fusion was associated with better clinical outcome scores or improvement in leg pain than nonunion. Crown Copyright Ó 2011 Published by Elsevier Inc. All rights reserved.
Keywords:
Radiographic solid fusion; Clinical outcome; Minimally invasive; TLIF
FDA device/drug status: not applicable. Author disclosures: none. * Corresponding author. Department of Orthopedic Surgery, Yonsei University College of Medicine, National Health Insurance Medical Center,
1232, Baeksok St, Ilsan district, Goyang city, Gyeonggi province, 410719, Republic of Korea. Tel.: (82) 31-900-0270; fax: (82) 31-900-0343. E-mail address: [email protected] (Y. Park)
1529-9430/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2011.01.023
206
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Context Several studies over the years have aimed to determine if the attainment of a demonstrably solid fusion correlates with clinical outcomes. Given the methodological limitations of many of these studies, the question remains open. In this article, the authors aim to gain some insight on this issue in patients undergoing transforaminal lumbar interbody fusion. Contribution The authors found that at midterm follow-up, patients who attained a solid fusion by their definition demonstrated improvements in leg pain and functional outcome similar to those who failed to fuse. However, the fusion group had greater improvement in back pain on VAS measurement. Paradoxically, the classification of ‘‘good-to-excellent results’’ was slightly better in the ‘‘nonunion’’ group. Implication The authors clearly outlined the limitations of the study in their discussion. The question addressed remains important in light of the fact that increasingly aggressive efforts are being made at achieving solid fusion, despite the fact that many randomized controlled trials (including SPORT) now suggest that more aggressive fusions do not necessarily improve outcomes as a whole for patients with degenerative indications. —The Editors
Introduction Minimally invasive lumbar fusion surgery has been reported to potentially offer a number of benefits over traditional open procedure, including less blood loss, less softtissue trauma, less postoperative back pain, shorter hospital stay, and earlier return to work [1–9]. In addition to these purported advantages, the achievement of a solid fusion is considered to be an ultimate surgical goal of minimally invasive spine fusion surgery as well as with open surgery. Many surgeons believe that patients achieving a radiographic solid fusion will exhibit more positive clinical outcome than those getting an unsuccessful radiographic fusion. Unlike other studies [10,11] that indicated no relation between radiographic solid fusion and clinical outcome, Kornblum et al. [12] found that successful fusion correlated with improved long-term clinical results in regard to back pain and lower limb function. To our knowledge, however, there is no study that has evaluated whether a radiographic solid fusion influences the clinical outcome of minimally invasive lumbar spinal
fusion. Our hypothesis is that a favorable clinical outcome is attained more often in patients achieving radiographic solid fusion rather than those failing to achieve radiographic fusion after minimally invasive spine fusion surgery. The purpose of this study was to evaluate the effect of radiographic solid fusion through minimally invasive transforaminal lumbar interbody fusion (TLIF) procedure on clinical outcome.
Materials and methods Patients’ population We performed a retrospective study by comparing the prospectively collecting data from consecutive patients who had achieved a solid fusion versus nonunion at least 2 years after minimally invasive TLIF by a single surgeon (YP) at an academic institution from 2004 to 2006. The surgical indications were patients who were suffering from low back pain and radiating pain down to lower extremity (leg pain) associated with a single-level lumbar or lumbosacral segmental instability: adult acquired lowgrade (Grade I/II) spondylolytic spondylolisthesis; Grade I/II degenerative spondylolisthesis with gross instability; and degenerative segmental instability combined with lumbar stenosis (central, lateral, and foraminal stenosis) and/or lumbar disc herniation. The inclusion criteria of segmental instability was $4 mm of translation or $10 of angular motion on preoperative flexion and extension radiographs. Our relative contraindications for minimally invasive approach, which were treated with traditional open surgery, included the following: high-grade (Grade III/IV) spondylolisthesis; patients with severely collapsed disc space and no motion on flexion-extension radiographs; patients who needed a multilevel decompression and fusion; those who had combined fixed coronal and/or sagittal deformities (kyphoscoliosis) that needed a surgical correction; or those who had a disease involving trauma, infection, and pathologic causes. All patients had preoperative evaluation with static and dynamic plain lumbar radiographs, magnetic resonance imaging, and/or computed tomography (CT). The surgery was undertaken when the preoperative symptoms and signs of back and leg pain were refractory to nonoperative treatment, such as medications, physiotherapy, epidural steroid injection, and/or a neurologic deterioration was developed. The surgical goal of all patients was a decompression and fusion for the treatment of back and leg pain arising from lumbar segmental instability. Surgical technique Under fluoroscopic guidance, an appropriate length of 22-mm–diameter METRx (Medtronic, Memphis, TN, USA) tubular retractor was introduced through a 2.5-cm incision for both neural decompression and access to the
Y. Park et al. / The Spine Journal 11 (2011) 205–212
interbody space. The approach was carried out on the side of the worst preoperative symptom of radiculopathy. Sextant screws and rod were placed for percutaneous pedicle screw fixation. A more detailed description of the procedure is available in the literatures [1–9]. In all cases, the autogenous bone obtained from the resected lamina and facet mixed with demineralized bone matrix (OsteofilRT DBM paste; Regeneration Technologies Inc., Alachua, FL, USA) was placed anteriorly and contralateral to the annulotomy within the interbody space, and then a polyetheretherketone cage (Capstone; Medtronic, Memphis, TN, USA) was inserted into the disc space. No additional contralateral facet fusion was performed in all patients. Data assessment Clinical outcome was evaluated using the patient-assessed quantitative measurement of visual analog scale (VAS) for back/leg pain down to lower extremity and Oswestry Disability Index (ODI), as well as the surgeon-assessed outcome measurement of a modified functional scale [13] (Table 1), both preoperatively and at each postoperative follow-up. The VAS scores were recorded on a 10-mm horizontal linewith 0 equal to ‘‘no pain’’ and 10 equal to ‘‘very severe pain.’’ The ODI was scored on a 0 to 100 scale using the Oswestry Disability Questionnaire. The subjective postoperative symptoms documented at each postoperative visit were divided into four categories: symptom free, back pain only, leg pain only, and both back and leg pain. Standing anteroposterior, lateral, flexion, and extension radiographs of the lumbosacral spine were collected from preoperative and final postoperative visit for the fusion assessment. Postoperative CT scan was also obtained at the final visit. All the clinical and radiographic data were reviewed by independent experienced spine surgeons (JWH, YTL) for analyses. A radiographic solid fusion was determined by following criteria that were proposed by Burkus et al. [14]; no motion (acceptable intraobserver measurement error was 3 angular motion or 3 mm of translation) on flexionextension lateral radiographs; continuous bony bridge within/around the cage (incorporation of the grafted bone into the vertebral end plates) on the CT scan (Figure, Top
207
Left, Top Right); lack of radiolucent lines around the graft and cage as well as absence of a lucent hollow around the pedicle screws on dynamic radiographs and/or CT scan; and new bone formation adjacent to or within the cage and/or fused posterior facet joint (the opposite side of TLIF approach) on the CT scan (Figure, Top Right, Bottom). The fusion status was subdivided into four grades: Grade 1 (definitely solid)—no motion on flexion-extension radiographs, continuous bony bridge within/around the cage, new bone formation adjacent to or within the cage, and/or fused posterior facet joint on CT scan; Grade 2 (possibly solid)—no motion on dynamic radiographs—and continuous bony incorporation within/around the cage, without evidence of new bone formation adjacent to or within the cage or facet joint fusion; Grade 3 (probably not solid)—no motion excluding evidence of bony incorporation within/around the cage; and Grade 4 (definitely not solid)—motion on dynamic radiographs with no evidence of bony bridge within/around the cage. According to our fusion grades, we considered the Grades 1 and 2 as a radiographic solid fusion and the Grades 3 and 4 as a nonunion (Table 2). Statistical methods Patients were divided into two groups: a radiographic solid fusion (control) group (Grades 1 and 2) and nonunion group (Grades 3 and 4). Comparison and correlation analyses were performed to evaluate the relationship between fusion status and clinical outcome. The demographics and characteristics and validated outcome measurements (such as VAS, ODI score, and functional scale) were compared between two groups. The comparisons between the groups were based on chi-square tests or Fisher exact tests for categorical variables, and on independent t tests for continuous variables. Paired t tests were used to evaluate the clinical improvement after surgery by analyses of changes over time of VAS and ODI scores. The phi coefficient was used for a measure of the association between two categorical variables, and the point-biserial correlation was also used to determine the correlation between a continuous variable and a categorical variable. In all analyses, a probability less than .05 considered to be statistically significant. The analyses were performed with the use of SPSS version 10.0 (SPSS Inc., Chicago, IL, USA).
Table 1 Criteria for the assessment of functional outcome Outcome
Pain
Medication
Activity
Work status
Excellent
None except for occasional back pain Markedly improved, occasional pain Some improvement No change in symptoms or a worsening of the patient’s condition
None
Normal
Normal
Occasional use of pain medication
Minimal functional limitations
Frequent use of pain medication Oral use of narcotics
Restricted Incapacitated
Return to work, although not at the same job activity Limited Disabled
Good Fair Poor
208
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Figure. A radiographic solid fusion demonstrated on sagittal and coronal images of computed tomography scan: (Top Left) continuous bony incorporation within and around cage; (Top Right) new bone formation at the posterior margin of interbody space; and (Bottom) fused posterior facet joint.
Results A total of 76 patients comprised the inclusion criteria of our study. Two patients died during the follow-up period without causes related to index surgery, and eight declined Table 2 Grades for the assessment of fusion status CT finding
Grade
No motion on flexionextension radiographs
Continuous bony incorporation within and/or around cage
New bone formation adjacent to or within the cage and/or fused posterior facet joint
Definitely solid Possibly solid Probably not solid Definitely not solid
Yes Yes Yes No
Yes Yes No No
Yes No No No
CT, computed tomography.
to participate the radiographic and CT evaluation at final follow-up. Finally, 66 patients (87%) completed the follow-up visit and were included in the analysis. There were 20 men and 46 (69.7%) women with an average age of 57.569.2 years (range, 40–81 years). The mean duration of follow-up was 36.169.9 months (range, 24–63 months). Thirteen patients (19.7%) were current smokers. Coexisting conditions included 24 (36.4%) hypertension, 10 (15.2%) diabetes, and 16 (24.2%) osteoporosis (t score #2.5). The average duration of preoperative symptoms was 2 years (range, 0.5–20 years). There were 24 degenerative spondylolisthesis, 23 spondylolytic spondylolisthesis, and 19 degenerative lumbar instabilities. The most common segment treated was L4–L5 (40/66, 61%), followed by L5–S1 (21/66, 32%), and then L3–L4 (5/66, 7%). There were no conversions to an open procedure. According to our criteria, the fusion status was demonstrated as following four grades: Grade 1 (definitely solid)
Y. Park et al. / The Spine Journal 11 (2011) 205–212
209
Table 3 Preoperative clinical features of nonunion group and controls who had a solid fusion Demographics and characteristics
Overall
Controls
Number of patients
66
51
15
Mean age (range) (y)
57.5 (40–81)
56.8 (40–81)
60.3 (45–74)
.19
Female gender (N)
46
36
10
.76y
Mean body mass index (range) (kg/m2) #25 (N) O25 (N)
24.6 (18.2–34.6) 37 29
24.4 (18.2–34.6) 29 22
25.2 (21.1–30.9) 8 7
.38 NSy
Mean height (range) (cm)
156.6 (142–178)
156 (142–176)
159.7 (145–178)
.15
ASA class 1/2/3/4 (N)
23/43/0/0
19/32/0/0
4/11/0/0
.65y
Patients who were smokers (N)
13
10
3
NSy
Patients who had osteoporosis (T#2.5) (N)
16
13
3
NSy
Mean duration of symptoms (range) (y)
2 (0.5–20)
1.8 (0.5–10)
2.8 (0.5–20)
.21
Mean periods of follow-up (range) (mo)
36.1 (24–63)
36.7 (24–63)
33.7 (24–52)
.31
Preoperative diagnosis (N) Spondylolytic spondylolisthesis (Meyerding Grade 1, Grade 2) Degenerative spondylolisthesis (Meyerding Grade 1, Grade 2) Degenerative lumbar instability Level of surgery (N) L3–L4 L4–L5 L5–S1
Nonunion
p Value*
.41y 23 (12, 11) 24 (22, 2) 19
17 (7, 10) 17 (15, 2) 17
6 (5, 1) 7 (7, 0) 2 .6y
5 40 21
3 32 16
2 8 5
N, number of patients; NS, not significant; ASA, American society of anesthesiologists. * The p values are based on chi-square or Fisher exact tests for categorical variables and on independent t tests for continuous variables. y The distribution of variables was not significantly different.
in 33 patients (50%); Grade 2 (possibly solid) in 18 (27.3%); Grade 3 (probably not solid) in 2 (3%); and Grade 4 (definitely not solid) in 13 (19.7%). On flexion-extension radiographs, no motion was demonstrated in 53 (80.3%) patients. A lucent hollow around the pedicle screws was found in 11 (84.6%) of 13 patients of Grade 4 and none of Grade 3. The final radiographic fusion rate of our series was 77% (51 patients of Grades 1 and 2). Nonunion was allocated to 15 patients (23%) of Grades 3 and 4. None of them undertook revision surgery to repair nonunion. Fifty-one patients comprised a solid fusion (control) group, and 15 patients were included in the nonunion group. The preoperative factors, such as age, gender, body mass index, general health status classified by American Society of Anesthesiologists, smoking, osteoporosis, preoperative diagnosis, and level of surgery, were similar between two groups. The detailed preoperative demographics and characteristics were illustrated in Table 3. The average pre- and postoperative final VAS (for back and leg pain) and ODI scores were comparable between the groups. Each preoperative VAS for back and leg pain and ODI scores were significantly improved at final postoperative follow-up in both groups (all measurements, p!.0001) (Table 4).
Forty-one patients (80%) in the control group and 13 (87%) in the nonunion group demonstrated an excellent or good result of the final functional scale [14]. The rate of solid fusion was 76% (41/54) in patients with excellent or good result and 83% (10/12) in patients with fair or poor result (statistically not significant) (Table 4). Twenty-five (49%) patients in the control group and eight (53%) in the nonunion group revealed none of the symptoms at final visit. Both groups revealed similar distributions of final functional outcome (excellent and good vs. fair and poor) and postoperative symptoms (symptom free or not) (Table 4). Nevertheless, the postoperative improvement of VAS for back pain was significantly higher (p5.04) in the control group (3.9, 95% confidence interval of the paired difference, 3.2–4.6) than in the nonunion group (2.5, 95% confidence interval of the paired difference, 1.6–3.4). On the other hand, the comparable improvement of VAS for leg pain and ODI score is demonstrated between the two groups (Table 4). The presence of radiographic solid fusion positively, but not strongly, correlated with a postoperative improvement (paired difference) of VAS for back pain (r50.255, p5.039). On the contrary, it did not correlate with the
210
Y. Park et al. / The Spine Journal 11 (2011) 205–212
Table 4 Clinical outcome of nonunion group and controls who had a solid fusion Fusion status Clinical outcome
Overall
Controls
Nonunion
Number of patients
66
51
15
VAS for back pain (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
6.261.9 (3–10) 2.662.1 (0–8) 3.662.4 3.0–4.2 !.0001
6.462.0 (3–10) 2.562.1 (0–8) 3.962.5 3.2–4.6 !.0001
5.361.6 (3–8) 2.962.3 (0–7) 2.561.6 1.6–3.4 !.0001
.06 .58 .04
VAS for leg pain (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
8.162.0 (3–10) 1.662.8 (0–9) 6.563.1 5.8–7.3 !.0001
8.162.1 (3–10) 1.662.9 (0–9) 6.563.2 5.9–7.4 !.0001
8.361.6 (5–10) 1.762.7 (0–7) 6.762.8 5.1–8.2 !.0001
.67 .93 .85
ODI score (range)y Preoperative Final follow-up Paired differences 95% CI of the difference p Valuez
60.2616.5 (24–88) 25.9617.9 (0–78) 34.3620.2 29.3–39.2 !.0001
60.7617.5 (24–88) 26.6619.1 (0–78) 34.1621.9 28.0–40.3 !.0001
58.3613.0 (40–82) 23.6613.3 (6–56) 34.7613.4 27.3–42.1 !.0001
.62 .58 .93
Final functional scale, N (%) Excellent Good Fair Poor
37 17 10 2
(56) (26) (15) (3)
28 13 8 2
(55) (25) (16) (4)
9 (60) 4 (27) 2 (13) 0
.86x,k
Final postoperative symptoms, N (%) Free of symptoms Back pain Leg pain Back and leg pain
33 16 4 13
(50) (24) (6) (20)
25 14 3 9
(49) (27) (6) (18)
8 2 1 4
.69x,k
(53) (13) (7) (27)
p Value*
VAS, visual analog scale; ODI, Oswestry Disability Index; CI, confidence interval. * The p values are based on independent t tests for continuous variables. y The values are given as the mean and the standard deviation. z The p values are based on paired t tests for the analyses of changes over time. x The p values are based on chi-square tests for categorical variables (excellent+good vs. fair+poor, free of symptoms vs. symptomatic). k The distribution of variables was not significantly different.
other variables of clinical outcome, such as final VAS and ODI score, postoperative symptoms, and functional scale (Table 5). The group with degenerative lumbar instabilities showed a higher fusion rate (89.5%) than another two groups with spondylolisthesis (73.9% in spondylolytic spondylolisthesis group and 70.8% in degenerative spondylolisthesis group, respectively). The difference, however, was not statistically significant (p5.35) in regard to fusion status among three different preoperative diagnoses (Table 6).
Discussion Minimally invasive TLIF procedure allows surgeon to preserve posterior spinal–stabilizing structures including lamina, unilateral facet joint, paraspinal muscles, and posterior ligaments [1–9]. For the reason that preserving these
structures can be helpful to sustain spinal stability even in the condition of pseudarthrosis, patients with radiographicfailed fusion after minimally invasive procedure may have not so much of a complaint in pain and disability as those having nonunion after open fusion surgery. Although the influence between fusion status and clinical outcome after traditional open posterior lumbar fusion has been examined extensively through clinical and radiographic studies [10–12,15–19], there are no series that have reported this topic after minimally invasive lumbar fusion, using multiple validated outcome instruments and multimodal radiographic assessment including high resolution CT scan. To assess the relationship of radiographic fusion with the clinical result, independent experienced spine surgeons examined a consecutive series of patients who had had minimally invasive TLIF performed by a single surgeon. In the present study, radiographic solid fusion was achieved in 77% of patients, and overall functional
Y. Park et al. / The Spine Journal 11 (2011) 205–212 Table 5 Correlations between fusion status and clinical outcome Fusion status
Clinical outcome
Correlations
The presence of solid fusion or not
VAS for final back pain
Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Point-biserial correlation p Value N Phi coefficient p Value N Phi coefficient p Value N
0.07 .576 66 0.255* .039 66 0.012 .926 66 0.024 .848 66 0.07 .576 66 0.011 .93 66 0.15 .743 66 0.069 .925 66
Paired difference of VAS for back pain VAS for final leg pain
Paired difference of VAS for leg pain Final ODI score
Paired difference of ODI score Final postoperative symptoms Final modified functional scale by Whitecloud et al. [13]
N, number of valid cases; VAS, visual analog scale; ODI, Oswestry Disability Index. * Correlation is significant at the 0.05 level (two tailed).
improvement was noted in 82% of patients after minimally invasive TLIF. Patients in both groups exhibited highly significant improvements in back pain and radiating pain down to lower extremity (leg pain) and disability compared with their preoperative baseline. Additionally, the present study demonstrates that patients who achieved solid fusion appear to have a greater improvement of back pain from baseline than those with nonunion, although it did not show a strong correlation. On the contrary, the improvement of leg pain and disability and functional outcome seem to be independent from the radiographic fusion. Table 6 Radiographic fusion status among three subgroups after minimally invasive TLIF Overall Number of patients 66
Group 1
Group 2
Group 3
23
24
19
p Value*
Fusion status Solid fusion, N (%) 51 (77.3) 17 (73.9) 15 (70.8) 17 (89.5) .35 Nonunion, N (%) 15 (22.7) 6 (26.1) 7 (29.2) 2 (10.5) TLIF, transforaminal lumbar interbody fusion. The values are given as the number of patients. Group 1: spondylolytic spondylolisthesis. Group 2: degenerative spondylolisthesis. Group 3: degenerative lumbar instability. * The p values are based on chi-square tests for the analyses of categorical variables.
211
Burkus et al. [14] suggested that new bone formation adjacent to or within the cage is the most reliable radiographic indication of a solid fusion. Additionally, they also proposed the other criteria substantiating a radiographic solid fusion: no significant motion at the instrumented spinal segment on dynamic studies; incorporation of grafted bone into the vertebral end plates; and absence of progressive radiographic lucent lines around the cage. Our inclusion criteria for a radiographic solid fusion contained these reliable radiographic clues. Although our fusion Grade 2 is literally ‘‘possibly solid,’’ and this grade does not include the criterion of new bone formation around the interbody space or facet joint fusion, this category obviously meets the other generally accepted criteria to substantiate a radiographic solid fusion as it mentioned as the above: no motion on flexion-extension lateral radiographs; continuous bony bridge within/around the cage on the CT scan; lack of radiolucent lines around the graft and cage; and absence of lucent hollow around the pedicle screws [14]. For that reason, we believe that our Grades 1 and 2 are considered as a radiographic solid fusion. On the contrary, fusion Grade 3 does not demonstrate any evidence of continuous bony incorporation within or around the cage but just grafted bone observed within the cage on the CT scan. Although the Grade 3 has no motion on dynamic radiographs at the instrumented spinal motion segment, this category corroborates nonunion. As we have used such a rigorous definition of fusion, the radiographic solid fusion was demonstrated only in 77% of our series. In the previous studies [2,5,7,9], almost 100% of fusion rate was reported based on the flexion-extension radiographs and/or postoperative CT scan at 2 years after minimally TLIF and significantly improved outcomes were also noted. In particular, recombinant human bone morphogenetic protein-2 soaked in a collagen sponge carrier was used as a graft material in these series. Direct comparison of our fusion rate with others appears to be inadequate because of the different graft material and methods of measurement. Application of rigorous fusion criteria and nonuse of recombinant human bone morphogenetic protein-2 might explain the relatively low rate of fusion compared with historical studies [2,5,7,9]. In the present study, postoperative back pain, leg pain, and disability were comparable in the radiographic solid fusion versus nonunion group. The patients with radiographic nonunion did not have so much of a complaint in pain and disability, rendering revision to repair nonunion not necessary. Although the final postoperative back pain scores were similar in both groups, considering the improvement scale from the preoperative to the postoperative period, a significantly better improvement was noted in the solid fusion group, whose correlation was not so strong statistically. It might be coming from one of the reasons that reducing iatrogenic soft-tissue damage by minimally invasive lumbar fusion procedure could contribute to decrease the back pain and disability of patients even
212
Y. Park et al. / The Spine Journal 11 (2011) 205–212
whose radiographic solid fusions were not achieved. However, this was not statistically confirmed by comparison with traditional open lumbar fusion in the present study, which is going to be the future topic of our study. As with other studies, the present study has several limitations. Although the clinical and radiographic data were collected prospectively, the present study is retrospective with a heterogeneous population, and the patients were not randomly selected. A more homogenous population might have provided a stronger correlation between independent variables and radiographic fusion status. In addition, the cohort size of the study is not large enough to have a sufficient statistical power for complete conclusions in some aspects. Another potential limitation of the study is that our average period of follow-up is not sufficiently long to validate the effect of minimally invasive lumbar fusion on the outcomes such as adjacent segment degeneration. Furthermore, given the rarity of minimally invasive spinal fusion surgeries performing at multiple levels, the data from multiple segmental lumbar fusions were not available for the assessment as well. Additionally, the present study did not examine the radiographic measurements to determine postoperative sagittal alignment, lumbar lordosis, and correction of a slip and collapsed disc space. Finally, our study has no control group for comparison. A matched cohort of patients underwent traditional open surgery that would provide the most ideal control group. We would then be able to assess the difference of long-term clinical and radiographic results between minimally invasive and traditional open procedures. Despite of these weaknesses, the present study offers insight into the treatment results of minimally invasive TLIF and estimates the rate of a radiographic solid fusion at minimum 2 years after minimally invasive TLIF. Although our findings could not identify risk factors associated with subsequent development of nonunion, our data provide important information regarding midterm results of minimally invasive TLIF. Our hypothesis is disputed against the results of the present study so that a favorable clinical outcome is not always more likely attained in patients achieving a radiographic solid fusion rather than those developing failed fusion after minimally invasive spine fusion surgery. As a result of the present study, after at least 2 years after minimally invasive TLIF, better reduction of back pain was noted in patients who achieved a radiographic solid fusion as opposed to those with nonunion. However, there was no clear evidence that radiographic solid fusion was associated with better clinical outcome scores or improvement of leg pain than nonunion. References [1] Dhall SS, Wang MY, Mummaneni PV. Clinical and radiographic comparison of mini-open transforaminal lumbar interbody fusion
[2] [3]
[4] [5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
with open transforaminal lumbar interbody fusion in 42 patients with long-term follow-up. J Neurosurg Spine 2008;9:560–5. Foley KT, Holly LT, Schwender JD. Minimally invasive lumbar fusion. Spine 2003;28(15 Suppl):S26–35. Isaacs RE, Podichetty VK, Santiago P, et al. Minimally invasive microendoscopy-assisted transforaminal lumbar interbody fusion with instrumentation. J Neurosurg Spine 2005;3:98–105. Mummaneni PV, Rodts GE Jr. The mini-open transforaminal lumbar interbody fusion. Neurosurgery 2005;57(4 Suppl):256–61. Park P, Foley KT. Minimally invasive transforaminal lumbar interbody fusion with reduction of spondylolisthesis: technique and outcomes after a minimum of 2 years’ follow-up. Neurosurg Focus 2008;25:1–7. Park Y, Ha JW. Comparison of one-level posterior lumbar interbody fusion performed with a minimally invasive approach or a traditional open approach. Spine 2007;32:537–43. Peng CWB, Yue WM, Poh SY, et al. Clinical and radiological outcomes of minimally invasive versus open transforaminal lumbar interbody fusion. Spine 2009;34:1385–9. Scheufler KM, Dohmen H, Vougioukas VI. Percutaneous transforaminal lumbar interbody fusion for the treatment of degenerative lumbar instability. Neurosurgery 2007;60(4 Suppl):203–12. Schwender J, Holly LT, Rouben DP, Foley KT. Minimally invasive transforaminal lumbar interbody fusion (TLIF). J Spinal Disord Tech 2005;18(Suppl):S1–6. Herkowitz HN, Kurz LT. Degenerative lumbar spondylolisthesis with spinal stenosis. A prospective study comparing decompression with decompression and intertransverse process arthrodesis. J Bone Joint Surg Am 1991;73:802–8. Fischgrund JS, Mackay M, Herkowitz HN, et al. 1997 Volvo Award winner in clinical studies. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective, randomized study comparing decompressive laminectomy and arthrodesis with and without spinal instrumentation. Spine 1997;22:2807–12. Kornblum MB, Fischgrund JS, Herkowitz HN, et al. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective longterm study comparing fusion and pseudarthrosis. Spine 2004;29: 726–33. Whitecloud TS, Davis JM, Olive PM. Operative treatment of the degenerated segment adjacent to a lumbar fusion. Spine 1994;19: 531–6. Burkus JK, Foley KT, Haid R, LeHuec JC. Surgical interbody research group—radiographic assessment of interbody fusion devices: fusion criteria for anterior lumbar interbody surgery. Neurosurg Focus 2001;10:E11. Christensen FB, Hansen ES, Eiskjær SP, et al. Circumferential lumbar spinal fusion with Brantigan cage versus posterolateral fusion with titanium Cotrel-Dubousset instrumentation: a prospective, randomized clinical study of 146 patients. Spine 2002;27: 2674–83. Diedrich O, Perlick L, Schmitt O, Kraft CN. Radiographic characteristics on conventional radiographs after posterior lumbar interbody fusion: comparative study between radiotranslucent and radiopaque cages. J Spinal Disord 2001;14:522–32. Greenough CG, Peterson MD, Hadlow S, Fraser RD. Instrumented posterolateral lumbar fusion. Results and comparison with anterior interbody fusion. Spine 1998;23:479–86. Wetzel FT, Brustein M, Phillips FM, Trott S. Hardware failure in an unconstrained lumbar pedicle screw system: a 2-year follow-up study. Spine 1999;24:1138–43. Resnick DK, Choudhri TF, Dailey AT, et al. Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 5: correlation between radiographic and functional outcome. J Neurosurg Spine 2005;2:658–61.