Background

TLIF is a refined surgical method that has evolved from the posterior lumbar interbody fusion (PLIF) technique. Blume and Rojas first proposed it in the early 1980s, and it has gradually replaced PLIF as a classic surgical approach for treating lumbar degenerative diseases [1]. The TLIF technique involves using a unilateral transforaminal approach, which avoids excessive traction of the dural sac, nerve roots, and back muscles. It achieves bilateral decompression by retaining the contralateral lamina and facet joints, with little impact on the mechanical structure of the spine’s posterior column. This technique also retains bone structures such as the pedicle and lamina, increasing the stability between the upper and lower adjacent vertebral bodies [2]. Long-term clinical studies have shown that TLIF has satisfactory clinical efficacy [3, 4]. However, despite its success, complications can arise during the procedure. A common complication is contralateral symptomatic FS after unilateral TLIF, which can negatively impact the operation’s comprehensive efficacy. Previous studies have identified several important risk factors for this complication, including preoperative contralateral intervertebral foramen stenosis, contralateral lateral recess stenosis, large lumbar sagittal plane mobility, and intraoperative fusion device position deviation to one side. Nonetheless, these studies have shortcomings, and the clinical significance of their findings is questionable [5, 6]. Therefore, we conducted a retrospective study of cases of unilateral TLIF in our hospital from January 2017 to January 2021. The study aimed to clarify the risk factors of nerve root symptoms due to contralateral FS after unilateral TLIF and to guide and standardize the operation process of unilateral TLIF to reduce the occurrence of postoperative contralateral symptomatic FS.

Information and methods

Inclusion and exclusion criteria

The inclusion criteria were low back pain with unilateral nerve root symptoms, surgery for unilateral TLIF, the surgical segment between L3 and S1, and lesions involving four segments below, and we excluded patients who had bilateral nerve root symptoms or severe systemic diseases like lumbar trauma, tumors, severe osteoporosis, birth defects or surgeries involving more than four segments, non-unilateral TLIF, or those lacking long-term complete follow-up and clinical data.

General information and case subgroups

In this study, 487 patients (269 males and 218 females) were included with a mean age of 57.1 years (range, 48–77 years). Prior to surgery, all patients underwent lumbar spine X-ray, CT plain scan, and lumbar MRI. Within one week after the operation, patients underwent a reexamination of lumbar spine X-ray and CT plain scan. Additional lumbar MRI was conducted on patients who experienced contralateral nerve root symptoms after surgery, and a nerve root block was performed if needed to identify the cause. Of the total patients, 32 experienced contralateral nerve root symptoms after the operation. The causes were confirmed by imaging and nerve root block examination: 23 cases were attributed to contralateral foraminal stenosis, three cases were due to improper screw position, four cases were caused by postoperative hematoma, and two cases had unknown causes. All 23 cases in which contralateral nerve root symptoms were caused by contralateral foraminal stenosis showed symptoms within one week after the operation, including lower limb pain, hypoaesthesia, paraesthesia, and decreased leg muscle strength. Sixty patients who did not experience contralateral nerve root symptoms after surgery were randomly selected and included in group B using a random number table method.

Among the 23 patients in group A who developed contralateral nerve root symptoms after surgery, immediate conservative measures, including dehydration and steroid therapy, were employed for a period of one to two weeks. Of these patients, 18 exhibited a gradual reduction and improvement of symptoms within two to three days of conservative treatment. However, five patients received revision surgery due to unclear treatment efficacy. The revision surgery was successful, and there was no deterioration of neurological symptoms or changes in muscle strength after surgery. The overall recovery rate was deemed acceptable.

Surgical operation and postoperative management

All patients in this study underwent more than three months of conservative treatment before surgery was considered. Unilateral TLIF was performed by two senior surgeons from the same institution (W-Y.J. and W–H.M.). The patients were positioned prone with abdominal suspension after being given general anaesthesia, and a longitudinal incision of 3–5 cm was made on the affected side of the spine to determine the surgical segment. The skin and fascia were then cut and separated to reveal the outer edge of the lamina and the upper and lower articular processes. On the decompression side, the ligamentum flavum, partial superior, and inferior articular processes were removed, with the extent of resection not exceeding 1/3 of the medial side. After careful preparation of the endplate, an autogenous bone-filling cage was inserted into the intervertebral space, followed by the insertion of the pedicle screw by hand. The pre-bending titanium rod was placed on the screw, and the screw rod was locked after moderate intervertebral compression. One drainage tube was retained, and the incision was closed in layers after wound haemostasis and irrigation.

Following anaesthesia recovery, patients received non-steroidal anti-inflammatory drugs (NSAIDs) and methylprednisolone to alleviate nerve root pain and inflammation. To prevent deep vein thrombosis, elastic socks and an arteriovenous foot pump were utilized six h after the operation. Once the drainage tube was removed, patients were required to walk with brace protection. In the event that patients experienced contralateral nerve root symptoms after the operation, timely administration of NSAIDs was given. If a single injection of 1% lidocaine 0.5 ml resulted in an improvement of root symptoms by ≥ 80%, the clinical diagnosis was nerve root lesions, and a lumbar MRI was performed to determine the cause.

Imaging evaluation

Within the first week post-operation, all patients underwent a lumbar spine X-ray, lumbar CT scan, and three-dimensional reconstruction to assess the screw position and rule out contralateral nerve root symptoms caused by improper screw placement. Patients with contralateral radicular symptoms underwent additional lumbar MRI to identify the cause of the disease, and a nerve root block was performed if needed to exclude symptoms caused by postoperative haematoma or new disc herniation. The evaluation of screw position was based on the standard of Liu Qingyu et al. [7], while postoperative haematoma was diagnosed according to the presence of an equal or slightly low signal shadow with progressive enlargement on the acute MRI T2WI image. The diagnosis of FS was based on the imaging diagnostic criteria of Widermuth et al. [8].

Measurement of contralateral foramen area CFA (Fig. 1)

Fig. 1
figure 1

a CFA measurement map of parasagittal MRI reconstruction (red F represents bone structure CFA, green E represents CFA considering soft tissue, F > E). b Lumbar lordosis angle LL: the angle between the upper endplates of L1 and S1. c Segment lordosis angle SL: the angle between the upper and lower endplates of the surgical segment. d Disc height DH: the distance between the upper and lower endplates of the intervertebral space center of the surgical segment; intervertebral space height FH: the distance between pedicles; intervertebral space width FW: The maximum width between the posterior edge of the vertebral body and the anterior edge of the superior articular process of the lower vertebral body

Full size image

CFA was measured using magnetic resonance imaging (MRI) T2-weighted sagittal plane images, and the measurements were obtained using the PACS 3.0 (In-finite, China) irregular surface measurement tool. To account for differences in intervertebral foramen dimensions across different sagittal planes, the paramedian sagittal plane, which can display the pedicle, as well as the superior and inferior articular processes simultaneously, was selected for intervertebral foramen measurement [9, 10].

Measurement of lumbar lordosis angle (LL) and segmental lordosis angle (SL)

LL and SL were assessed both preoperatively and postoperatively. LL was measured as the angle formed by the upper endplates of L1 and S1, while SL was measured using the Cobb method, which involved measuring the angle formed by the vertical lines of the upper and lower endplates of the surgical segment.

Measurement of disc height (DH), foramen height (FH), and foramen width (FW)

DH, FH, and FW were measured using CT sagittal images. DH was defined as the distance between the centres of the upper and lower endplates in the median sagittal plane. FH was measured as the distance between the pedicles, while FW was defined as the maximum width between the posterior edge of the vertebral body and the anterior edge of the superior facet of the lower vertebral body.

Description of the position of the fusion cage (Fig. 2)

Fig. 2
figure 2

a The center of the marker fuser, i.e., the centre of gravity of the triangle formed by the three marker points of the fuser. b Measure the distance A from the centre of the cage to the anterior edge of the vertebral body, measure the sagittal diameter length B of the intervertebral disc, and A/B is the sagittal position of the cage. c The distance C from the center of the fusion cage to the coronal bisector of the intervertebral disc is negative when the centre of the fusion cage is deviated from the insertion side of the fusion cage and positive when the centre of the fusion cage is deviated to the opposite side. In this case, the fusion cage is inserted from the left side, so C is positive; left and right disc diameter length D; C/D is the coronal position of the fusion cage

Full size image

The position of the fusion cage was assessed using cross-sectional CT scans. The centre of the cage was marked, which was the centre of gravity of the triangle formed by three marked points of the cage (Fig. 2a). The distance from the centre of the cage to the vertebral body was measured as A, and the length of the sagittal diameter of the intervertebral disc was measured as B. The ratio of A to B was calculated to determine the sagittal position of the cage (Fig. 2b). The distance from the centre of the cage to the bisector of the coronal plane of the intervertebral disc was measured as C, which had both positive and negative values. A negative value indicated that the centre of the cage was biased toward the side of the cage, whereas a positive value indicated that it was biased toward the opposite side. The length of the left and right diameters of the intervertebral disc was measured as D, and the ratio of C to D was calculated to determine the coronal position of the cage (Fig. 2c) [11].

The variation before and after the operation was calculated by subtracting the preoperative values from the postoperative values for all the parameters mentioned above. The measurements of all these parameters were performed by a spinal surgeon with over ten years of experience in spinal surgery and a radiologist specialized in musculoskeletal systems. The final results were obtained by averaging two measurements.

Clinical efficacy evaluation

The patients were followed up at one, three, six and 12 months after the operation. The clinical efficacy of the two groups was evaluated using the VAS and the JOA lumbar score. The VAS score and JOA lumbar score were assessed immediately before the operation and at one year postoperatively.

Statistical analysis

All statistical analyses were conducted using SPSS 25.0 software. The normally distributed measurement data were presented as mean ± standard deviation (X ± S), and the independent sample t-test was utilized for analysis. Non-normally distributed measurement data were presented as medians (range from minimum to maximum), and the Mann–Whitney U test was employed. Count data were analyzed using the chi-square (X2) test. One-way ANOVA was employed to compare the mean difference of each index in different periods, and the LSD method was used for comparison between groups. Univariate intergroup analysis was conducted to screen variables, and logistic regression was utilized to analyze the risk factors of nerve root symptoms caused by contralateral FS after unilateral TLIF. Statistical significance was defined as P < 0.05.

Results

General data comparison

A total of 23 patients (14 males and 9 females) who experienced contralateral nerve root symptoms caused by contralateral FS were included in group A, while 60 patients without such symptoms were included in group B (38 males and 22 females). The mean age of group A was 55.8 ± 12.7 years, while that of group B was 53.1 ± 15.1 years. The incidence rate of contralateral nerve root symptoms caused by contralateral FS after unilateral TLIF was found to be 4.72% (23/487). There were no significant differences in gender, age, BMI, BMD, and diagnosis between the two groups, as shown in Table 1.

Table 1 Comparison of general data of patients with symptomatic spinal stenosis after unilateral TLIF
Full size table

Surgical related indicators

In this study, there was no statistically significant difference (P > 0.05) observed between the two groups in terms of the number of surgical segments involved (1, 2, or 3), the surgical levels involved (L3/4, L4/5, L5/S1, L3/5, L4/S1, L3/S1), and the height of the interbody fusion device (Table 2).

Table 2 Comparison of differences in surgical-related indicators between the two groups
Full size table

Imaging measurement

The study results indicated that there were no significant differences in LL, DH, and FH between the two groups before and after the operation. However, there was a statistically significant difference in preoperative CFA between group A and group B, with group A having a lower CFA (56.1 ± 15.2) compared to group B (67.1 ± 11.9). Additionally, the difference in postoperative CFA between the two groups was greater, with group A having a lower CFA(42.3 ± 12.0) compared to group B(75.4 ± 14.8), and the change in CFA was less in group A (− 13.8 ± 13.2) compared to group B (8.3 ± 15.2), which was statistically significant (P < 0.001). The CFA of group A showed a downward trend, while that of group B showed an upward trend. The difference in SL between the two groups before and after the operation was statistically significant (P < 0.05), with values of 10.7 ± 9.8 and 15.6 ± 5.1, 6.3 ± 2.9 and 1.8 ± 2.2, and 21.3 ± 7.9 and 16.7 ± 5.2 for group A and group B, respectively. There were also significant differences in FW between the two groups before and after the operation (P < 0.05), with values of 8.1 ± 3.2 and 9.3 ± 1.1, 5.2 ± 3.3 and 8.7 ± 2.0, and 3.6 ± 1.4 and 1.2 ± 1.6 for group A and group B, respectively. No statistically significant difference was found in the sagittal position of the fusion cage between the two groups (44.5 ± 6.8% for group A and 47.2 ± 5.7% for group B), but there was a significant difference in the coronal position (− 6.5 ± 7.8 for group A and 8.2 ± 4.1 for group B) (Table 3).

Table 3 Comparison of imaging parameters in patients with symptomatic spinal stenosis after unilateral TLIF (x ± s)
Full size table

Clinical observation indicators

The statistical analysis of the preoperative and postoperative VAS and JOA scores between the two groups revealed no significant difference in the preoperative scores (P = 0.351). However, the pain VAS score one year after the operation was significantly lower than that before the operation, and the JOA lumbar spine score was significantly higher than that before the operation (P < 0.05). There was no significant difference in VAS scores between the two groups (P = 0.483). The JOA lumbar score in group B was significantly higher than that in group A (P < 0.001, Table 4).

Table 4 Comparison of clinical scores of patients with symptomatic spinal stenosis after unilateral TLIF (x ± s)
Full size table

Logistics regression analysis

Four variables with significant differences between the two groups in the single-factor group analysis were included in the logistic regression analysis, and the variables were screened backward stepwise (Wald). The results showed that preoperative contralateral foramen area (OR = 1.176, 95% CI: 1.012–1.367), segmental lordosis angle (OR = 2.225, 95% CI: 1.124–4.406), contralateral intervertebral foramen width (OR = 2.706, 95% CI: 1.028–7.118), and fusion cage coronal position not crossing the midline (OR = 1.567, 95% CI: 1.142–2.149) were independent risk factors for contralateral symptomatic FS after unilateral TLIF (refer to Figs. 3 and 4). Figures 5 and 6 depict typical cases.

Fig. 3
figure 3

Regression analysis results of contralateral symptomatic FS logistics after unilateral TLIF. Note: P < 0.05 was considered statistically significant. Hossmer-Lemeshaw test P = 0.986, fitting well

Full size image
Fig. 4
figure 4

Forest map of the results of contralateral symptomatic FS logistics regression analysis after unilateral TLIF

Full size image
Fig. 5
figure 5

A 53-year-old male patient with low back pain, intermittent claudication of the right lower limb, numbness, pain, and weakness for more than 1 year aggravated for 1 week. a, b L5/S1 bilateral lumbar intervertebral foramen stenosis, severe stenosis on the right side, and mild stenosis on the left side. c The positive and lateral X-rays of the lumbar spine were reviewed 1 month after the operation, and the decompression and interbody fusion were performed through the right intervertebral foramen of L5/S1. The left side did not undergo preventive decompression, and the screws were in place. d One year after lumbar CT review, good intervertebral fusion, fusion position in position (sagittal position slightly ahead of the coronal position according to the middle of the intervertebral space), no endplate collapse, and fusion slip occur

Full size image
Fig. 6
figure 6

A 62-year-old female patient with low back pain and numbness of the right lower limb for more than 20 years was admitted to the hospital. Combined with the patient’s symptoms, signs, and imaging data, lumbar decompression and interbody fusion were performed through the L3/4 right intervertebral foramen, and no preventive decompression was performed on the left side during the operation. Two days after the operation, the patient developed numbness in the contralateral (left) lower limb, and the anterior tibial muscle strength was grade 0. Conservative treatments such as anti-inflammatory medication, nerve repair, and anti-inflammatory analgesia were given. On the third day, the patient’s symptoms improved. a Preoperative MRI L3/4 disc herniation, spinal stenosis. b Lumbar MRI 2 days after operation showed no contralateral recess stenosis, disc herniation, hematoma, etc. c One year after the operation, lumbar CT showed good intervertebral fusion, the position of the fusion cage was in place, no endplate collapse, fusion cage slip, etc. d Patients after a 1-month review; screws in place; no broken rods; and no broken nails

Full size image

Discussion

TLIF is a surgical procedure that can provide bilateral decompression and sufficient biomechanical stability to promote intervertebral fusion by reaching the intervertebral space through a unilateral intervertebral foramen. TLIF is a widely used technique with excellent curative effects; however, contralateral symptomatic FS often occurs after a unilateral approach. Previous studies have reported the incidence of contralateral root symptoms after microendoscopy-assisted minimally invasive TLIF to be 8.5%, and after open TLIF, it is 4.78%, which is similar to the incidence of 4.72% in this study [12].

Since Hunt et al.’s report of a case of contralateral root symptoms after unilateral TLIF in 2007, several scholars have conducted in-depth research on its incidence and risk factors. Yang et al. [13] found that preoperative contralateral FS was the most common cause of postoperative contralateral symptomatic FS. In the L4-5 segment, the optimal threshold of preoperative CFA was 0.76 cm2. Cho et al. [14], after studying the morphology of the intervertebral foramen in 33 patients with unilateral TLIF, proposed that the position of the implanted cage played a decisive role in the change in the morphology of the intervertebral foramen. Hwang et al. [12] conducted a retrospective study of 592 patients undergoing unilateral TLIF and found that the preoperative sagittal range of motion (OR = 1.562; P = 0.004) and fuser position (OR = 2.047; P = 0.015) were associated with contralateral root symptoms after unilateral TLIF. However, these studies have some limitations. For example, the research data of Hunt et al. were not measured in the same CT sagittal plane, which raises concerns about the reliability of the data. Yang Y et al.’s study was based on L4/5 segment CT image measurement data without taking into account the ligament fat within the intervertebral foramen. Moreover, the lumbar lateral flexion radiographs involved in Hwang et al.’s study are not common and have limited clinical significance. To address these limitations, we combined the anatomical study of the intervertebral foramen [15, 16] and changed the measurement of the CFA from the CT sagittal plane to the MRI (T2W1) sagittal plane. Furthermore, we defined the FW as the maximum width between the posterior edge of the vertebral body and the anterior edge of the superior articular process of the lower vertebral body. The position of the cage is described by the coronal and sagittal plane positions, and the previous experimental research is adjusted and improved to clarify its risk factors and guide clinical prevention and treatment.

Relationship between preoperative contralateral foramen stenosis and contralateral symptomatic FS after unilateral TLIF

Several studies have highlighted the significance of decompressing the intervertebral foramen nerve root during TLIF to improve symptoms [17, 18]. In cases where extensive decompression is unnecessary, bilateral approach decompression can be avoided, and indirect decompression can be used to alleviate contralateral FS. This involves implanting an appropriate cage to increase CFA. However, clinical experience shows that CFA does not always increase as expected. In this study, the preoperative and postoperative CFA of group A was smaller than that of group B, and the difference was statistically significant. Additionally, the CFA of group A showed a downward trend, while that of group B showed an upward trend. Logistic analysis indicated that preoperative contralateral foramen stenosis was an independent risk factor for postoperative contralateral symptomatic FS.

Relationship between preoperative segmental lordosis angle and contralateral symptomatic FS after unilateral TLIF

The appropriate recovery of lumbar lordosis in lumbar surgery is crucial for improving clinical outcomes and preventing congenital flat back [19]. However, in this study, there were no significant differences in terms of the number of surgical segments, the surgical levels and lumbar lordosis angle LL (preoperative, postoperative, and change) between the two groups, but there were statistical differences in the preoperative, postoperative, and change of segmental lordosis angle SL. Logistic analysis revealed that the preoperative segmental lordosis angle was an independent risk factor for postoperative contralateral symptomatic FS. These results are similar to those reported by Jiang et al. [20], suggesting that excessive pursuit of recovery of lumbar lordosis may contribute to this outcome. Unilateral TLIF only removes one facet joint, and the posterior structure of the lumbar spine experiences less damage. However, the sagittal position of the fusion cage tends to be biased toward the middle and posterior sides, resulting in the hinge fulcrum being moved backward and the force arm becoming smaller, which requires greater stress to recover the lordosis angle. During this process, careless operation may cause the superior articular process of the lower vertebral body to move upward or ventrally, leading to the dislocation of the superior articular process and compression of the intervertebral foramen nerve root.

Relationship between preoperative contralateral intervertebral foramen width and contralateral symptomatic FS after unilateral TLIF

Studies have indicated that the reduction of the intervertebral foramen width FW is more likely to cause intervertebral foramen nerve root compression than the intervertebral foramen height FH. Intervertebral foramen posterior wall degeneration of hypertrophic ligaments, joint capsule, and the proliferation of facet joints are common causes of nerve root compression [21, 22]. In this study, there were statistical differences in the preoperative and postoperative periods and changes in the width of FW between the two groups. Logistic analysis revealed that a smaller width of the contralateral intervertebral foramen before surgery was an independent risk factor for postoperative symptomatic FS. During the operation, when the screw rod is locked, the superior facet of the lower vertebral body tends to move upward or ventrally. Compared to the intervertebral foramen height, the nerve root has enough buffer space. However, even a 1–2 mm reduction in the width of the intervertebral foramen can have a significant impact.

Relationship between fusion cage and contralateral symptomatic FS after unilateral TLIF

Clinical studies have demonstrated that intervertebral space and foramen height may not always recover effectively after unilateral TLIF, particularly on the uncompressed side. This may lead to new neurological symptoms [23]. The authors suggest that this could be due to the fusion implant position being biased toward the decompression side. In this study, there was no significant difference in cage height and the sagittal position of the cages between the two groups, but there was a statistical difference in the coronal position. Logistic analysis revealed that the coronal position of the cage, which did not cross the midline, was an independent risk factor for contralateral symptomatic FS after unilateral TLIF. Prior studies have indicated that when the cage height is less than the preoperative intervertebral space height, it is easier to achieve the required SL during surgery, but there is a higher risk of cage slip. Conversely, when the cage height is too high, it is difficult to obtain the required SL during the operation, but slipping is unlikely to occur. Therefore, there is a correlation between cage height and SL [24]. However, in our study, we kept the cage height consistent with DH, avoiding this issue. Thus, cage height did not affect the recovery of SL and was not associated with postoperative symptomatic intervertebral foramen stenosis. Furthermore, when the coronal position of the cage is biased toward the decompression side, the stress distribution inside the intervertebral space becomes unbalanced, causing the intervertebral space to tilt. Consequently, the height of the intervertebral foramen on the decompression side significantly increases, while the height of the intervertebral foramen on the non-decompression side decreases, leading to nerve root compression. A significant amount of bone grafting in the anterior part of the vertebral body near the anterior longitudinal ligament is favorable for vertebral bone fusion. Therefore, the sagittal position of the fusion cage should be located in the middle and posterior sides.

Conclusion

The incidence of contralateral symptomatic FS after unilateral TLIF is relatively low at 4.72%. However, this study found that certain preoperative factors such as contralateral intervertebral foramen stenosis, a smaller segmental lordosis angle, a smaller intervertebral foramen width, and the coronal position of the fusion cage not crossing the midline were independent risk factors for this complication. As a result, for patients with these risk factors, special attention should be given to the locking of the screw rod during lumbar lordosis recovery and the placement of the fusion cage to ensure proper coronal positioning beyond the midline. Additionally, prophylactic decompression may be necessary. It is important to note that this study has some limitations. For instance, it is a single-center study and may not have accounted for other potential risk factors. Furthermore, this study did not provide a quantitative analysis of imaging data, making it difficult to determine when preventive decompression should be performed. Therefore, future multi-centre, large-sample, and high-quality research is necessary to further improve our understanding of the risk factors and appropriate management of contralateral symptomatic FS after unilateral TLIF.