The traditional open transforaminal lumbar interbody fusion (TLIF) is successful in treating various symptomatic spinal conditions, but complications like extensive blood loss and prolonged recovery time continue to remain a major concern [
One major concern for MIS-TLIF is that it relies on fluoroscopic guidance rather than direct visualization of anatomic landmarks to insert pedicle screws. This loss of visualization increases the risk of facet join violation (FV) at the superior motion segment [
Several studies have reported the risk factors such as age, sex, weight, top-screw level in FV in percutaneous pedicle screw placement [
This study was reviewed and approved by the Human Ethics Committee of the Tongji Hospital, Tongji University School of Medicine, China. This was a retrospective study, so individual consent was waived.
A cohort of 99 adult patients (51 male, 48 female, average age:
Computed tomography (CT) and magnetic resonance imaging (MRI) were used to classify lumbar disc herniation, lumbar spinal stenosis, and spondylolisthesis. All operations were performed by 3 experienced surgeons who had over 10 years’ experience in spine surgery.
Radiographic evaluation of top level screws, facet grade, adjacent segment disc height, Cobb angle, and other parameters was made by 2 different radiologists and spine surgeons for an accurate analysis. Shah’s classification was used to identify FVs in patients at 1-year follow-up [
General information of different FV patients in MIS-TLIF.
Group A (31 patients) | Group B (22 patients) | | |
---|---|---|---|
| 55.23 ± 12.09 | 52.23 ± 12.65 | 0.387 |
| 0.833 | ||
Male | 15 | 10 | |
Female | 16 | 12 | |
| 27.42 ± 1.83 | 28.51 ± 3.13 | 0.150 |
| 0.828 | ||
L4 | 22 | 15 | |
L5 | 9 | 7 | |
| 18.26 ± 3.78 | 18.59 ± 4.24 | 0.765 |
BMI: body mass index.
Patients were kept in prone position following general anesthesia and their abdomen was suspended and pressure parts were with pad. C-arm fluoroscopy was used to determine available surgical space. A 2 to 3 cm incision was made approximately 2.5 cm lateral to the midline to cut skin and muscular fasciae. After inserting the dilators step by step, Pipeline working channel (Johnson & Johnson Company, NY, USA) was placed into and fixed by dilators, or directly using Spotlight working channel (Johnson & Johnson Company, NY, USA). Then the local soft tissue was removed to expose vertebral plate edges and facet joint. The decompression was performed to expose dural sac, the central canal, lateral crypt, and nerve root canal after removing part of vertebral plate, ligamentum flavum, and facet joints. After thoroughly removing intervertebral disc and cartilage endplate, local autologous bone was implanted into intervertebral space, and then single suitable height of intervertebral fusion was placed. For bilateral decompression or more, the same method was performed to deal with the contralateral and other spaces. Under the guidance of C-arm fluoroscopy, placement of percutaneous pedicle screws was performed using Viper 2 system (Johnson & Johnson Company, NY, USA), and percutaneous rod was also placed using the instruments and prelocked. Drainage tube was removed 24–36 h postoperatively. At 3 days postoperatively, the patients were examined with lumbar X-ray and CT to confirm the position of lumbar fusion instruments and internal fixation and to evaluate the facet joint violation. Patients were encouraged to have activities out of bed under waist protection. Waist torsion and bending activities were prohibited within 3 months under waist protection.
High-resolution CT scan with sagittal and coronal reconstructive images were adjusted parallel to the pedicle screws to evaluate the positions in all patients. The facet joint violations were examined using Shah’s classification (3 grades, Figure
Shah’s classification. (a) (grade 1): no facet joint violation; (b) (grade 2): unilateral facet joint violation (screw head contacted or invaded facet joint); (c) (grade 3): bilateral facet joint violation (screw head contacted or invaded facet joint).
Clinical data were collected to compare therapeutic efficacy among different FV patients. Operation duration, blood loss, and postoperative drainage were recorded. Average intervertebral disc height and lumbar and surgical Cobb angle were measured in X-rays before and after operation to assess the recovery of intervertebral space height and the variation of lumbar spine kyphosis (Figure
Radiographic method. (a) Average intervertebral disc height (AIDH): (1) anterior of intervertebral disc height, (2) middle of intervertebral disc height, (3) the posterior of intervertebral disc height, AIDH: ((1) + (2) + (3))/3; (b) adjacent superior average intervertebral disc height (ASAIDH) and adjacent superior intervertebral disc Cobb angle (ASIDCA): (4) anterior of adjacent superior intervertebral disc height, (5) middle of adjacent superior intervertebral disc height, (6) posterior of adjacent superior intervertebral disc height, ASAIDH: ((4) + (5) +(6))/3, ∠D for ASIDCA: angle of upper and low endplate; (c) lumbar sagittal curvature: ∠A represents lumbar Cobb: angle of L1 upper endplate and S1 upper endplate, ∠B for the L4/5 Cobb: angle of L4 upper endplate and L5 low endplate, ∠C for L5/S1 Cobb: angle of L5 upper endplate and S1 upper endplate.
All data were expressed as means ± standard deviation. Chi-squared analysis was carried out to determine the association between superior facet joint violation and technical factors. Logistics regression analysis was used to examine simultaneous relationships of violation’s risk. Student's
99 cases (total 198 top-screws) were operated successfully. Postoperative CT scans showed that the level of FV included 60 cases with grade 1 FV (60.61%), 32 cases with grade 2 FV (32.32%), and 7 cases with grade 3 FV (7.07%), and the incidence of FV in MIS-TLIF was 39.39% (39/99) of the patients and 23.23% (46/198) of the screws. Chi-square test showed that screw pedicle placement at L5 and patients with BMI > 30 kg/m2 had a higher prevalence rate of FV (Table
Relationship between FV and relevant factors.
Factors | Patient ( | Relationship between facet joint and patient | | |
---|---|---|---|---|
Grade 1 | Grade 2 + 3 | |||
| 0.653 | |||
Male | 51 | 32 | 19 (grade 2: 15, grade 3: 4) | |
Female | 48 | 28 | 20 (grade 2: 17, grade 3: 3) | |
| 0.566 | |||
<60 | 60 | 35 | 25 (grade 2: 21, grade 3: 4) | |
>60 | 39 | 25 | 14 (grade 2: 11, grade 3: 3) | |
| <0.00 | |||
<30 kg/m2 | 62 | 46 | 16 (grade 2: 14, grade 3: 2) | |
>30 kg/m2 | 37 | 14 | 23 (grade 2: 18, grade 3: 5) | |
| | |||
L3 | 4 | 4 | 0 (grade 2: 0, grade 3: 0) | |
L4 | 66 | 44 | 22 (grade 2: 21, grade 3: 1) | |
L5 | 29 | 12 | 17 (grade 2: 11, grade 3: 6) | |
| 0.864 | |||
Unilateral decompression | 87 | 53 | 34 (grade 2: 27, grade 3: 7) | |
Bilateral decompression | 12 | 7 | 5 (grade 2: 5, grade 3: 0) |
BMI: body mass index;
At 1-year follow-up, 59 cases with noninjury were assigned as group A. The average operative time was
Radiological and functional comparison between two groups.
Group A | Group B | | |
---|---|---|---|
AIDH (cm) | |||
Preoperation | 0.92 ± 0.28 | 0.92 ± 0.25 | 0.987 |
Final follow-up | 1.10 ± 0.23 | 1.12 ± 0.19 | 0.797 |
Lumbar Cobb angle | |||
Preoperation | 30.69 ± 11.60 | 33.32 ± 13.14 | 0.446 |
Final follow-up | 31.79 ± 12.54 | 35.30 ± 9.22 | 0.269 |
Surgical Cobb angle | |||
Preoperation | 14.13 ± 6.41 | 14.29 ± 6.63 | 0.931 |
Final follow-up | 13.84 ± 5.76 | 15.05 ± 4.66 | 0.422 |
Low back VAS | |||
Preoperation | 5.94 ± 1.03 | 6.23 ± 1.07 | 0.325 |
2-week postoperation | 3.13 ± 0.56 | 3.23 ± 0.43 | 0.494 |
Final follow-up | 1.48 ± 0.51 | 2.32 ± 0.72 | <0.001 |
Leg VAS | |||
Preoperation | 6.71 ± 1.01 | 6.86 ± 0.89 | 0.568 |
2-week postoperation | 2.77 ± 0.62 | 2.86 ± 0.71 | 0.627 |
Final follow-up | 1.45 ± 0.51 | 1.50 ± 0.51 | 0.734 |
ODI | |||
Preoperation | 61.26 ± 11.17 | 57.73 ± 12.74 | 0.290 |
Final follow-up | 15.06 ± 3.92 | 19.05 ± 5.30 | 0.003 |
ASAIDH (cm) | |||
3-day postoperation | 1.07 ± 0.22 | 1.05 ± 0.23 | 0.730 |
Final follow-up | 0.92 ± 0.23 | 0.91 ± 0.24 | 0.872 |
ASIDCA (°) | |||
3-day postoperation | 7.07 ± 2.30 | 7.11 ± 2.78 | 0.955 |
Final follow-up | 7.91 ± 2.63 | 8.76 ± 2.50 | 0.238 |
AIDH: average intervertebral disc height; VAS: visual analogue scale; ODI: Oswestry disability index; ASAIDH: adjacent superior average intervertebral disc height; ASIDCA: adjacent superior intervertebral disc Cobb angle.
Spinal fusion technique is the gold standard in treating spinal disorders but postoperation complications still remain a significant concern for surgeons. The intervertebral fusion technique alters the disc biomechanical environment, thus altering the motion and function of the index and adjacent segments. Specifically, biomechanical studies reveal increased intradiscal pressure, increased tensile and shearing stresses, and increased facet loads, accelerating degenerative changes [
Physical conditions play an important role in the development of FV and ASD during MIS-TILF. Babu et al. [
Most studies suggested that FV occurred in L4 and L5 more often than L1, L2, and L3. Anatomical studies confirm that the direction of facet joint articular surfaces transferred from sagittal to coronal plane in lumbar vertebrae [
Park et al. [
This study is limited by several reasons. First, the overall sample size in this study is small. Specifically, the number of patients with grade 3 FV needs a larger sample group to reach a normal distribution and statistical reliability. The follow-up duration was short and radiographic methods were simple, which could potentially leave out explanations for our results. Lastly, biomechanical and kinematic studies should be done to determine the etiology of FVs.
We conclude that MIS-TLIF is a feasible technique to solve the lumbar disc degenerative disease effectively in lumbar spine, although FVs still occur commonly during follow-up and ultimately resulted in ASD. Future investigations regarding better surgical techniques based on MIS-TLIF need to be conducted to reduce the incidence of FV and consequent ASD.
The authors declare that they have no conflicts of interest.
Co-first authors, Long Jia and Yan Yu, contributed equally to this work.
The study was supported by research grants from Shanghai Hospital Development Center (no. SHDC-12013108), the National Natural Science Foundation of China (no. 31300779), and the Shanghai Rising-Star Program (no. 18QA1403800).