Lamina Abrasion Caused by High-Speed Drill in Unilateral Door Cervical Laminoplasty: A Clinical Investigation

Background

Unilateral door cervical laminoplasty (UDCL) is a widely accepted treatment for severe multisegmented cervical spinal stenosis, and mini-plate fixation is currently recognized and used in UDCL [1,2]. Excellent biocompatibility, biomechanical performance, and significant treatment effect have been proved by finite element analysis [3] and many clinical studies [4,5].

During UDCL, the increases in sagittal canal diameter (SCD) and the cross-sectional area (CSA) of the spinal canal were determined by laminar opening extent (LOE), showing an insufficient LOE would lead to insufficient decompression of the spinal cord, while an excessive LOE is a risk factor for complications such as postoperative C5 palsy [6,7]. Moreover, according to LOE, a suitable size of mini-plate should be selected, which is closely related to the stability of the reconstructed cervical spine based on finite element analysis [8]. Therefore, an appropriate LOE as well as a suitable mini-plate size is particularly important for desirable outcomes.

Numerous studies continued to try to improve UDCL, and assessed the association of LOE with SCD and CSA to provide assistance for surgery; however, these findings were too limited to be widely adopted in clinical practice [9,10]. Moreover, the lamina abrasion caused by high-speed drill in UDCL has been neglected in these reports, which could lead to unreliable results [11–13]. The present study aims to develop the concept of effective laminar opening extent (ELOE with consideration of the lamina abrasion extracted by the drill bit in UDCL, and the relationships between ELOE and the SCD as well as the CSA of the spinal canal were analyzed. The results could provide references for the selection of optimally-sized mini-plates prior to surgery and estimate the SCD/CSA increase after surgery, thus further improving the effectiveness and reliability of UDCL.

Material and Methods

SUBJECTS:

We included patients who were previously diagnosed as having cervical spondylosis myelopathy and ossification of the posterior longitudinal ligament, and radiographically confirmed by a spine expert in our institution. The inclusion criteria were as follows: (a) age 18–85 years; (b) magnetic resonance imaging (MRI) showed spinal stenosis or cervical disc herniation involving ≥3 levels and spinal cord compression; (c) patients who have received conservative therapy for at least 3 months before surgery, but without satisfactory effect. The exclusion criteria were as follows: (a) patients with cervical trauma history, cervical deformity, or cervical tumors; (b) patients who underwent posterior-anterior cervical surgery or revision surgery. All participants were instructed to fill out the cervical Japanese Orthopaedic Association (JOA) questionnaire before and after surgery. The present study was approved in advance by the Ethics Committee of our institution (No. XJCR2019-19), and written informed consent was obtained from all participants and their families.

SURGICAL PROCEDURES:

A standard posterior midline approach was performed in each patient. The paravertebral muscles were detached, then we removed the spinous processes. To expand the central spinal canal and relieve compression, the UDCL was performed using a high-speed 2-mm drill according to the modified Hirabayashi method [14] between C3–7 vertebrae. For all patients, we used the left side as the opening side and the lamina was completely cut by the drill; subsequently, the left side was partially cut and formed a hinge. After the left side-opening procedure, the Centerpiece mini-plates (Medtronic, Inc., USA) with mini-screws were used for internal fixation at the opened space. All surgeries were carried out by a same surgery team.

RADIOLOGY:

Computed tomography (CT) scans and three-dimensional reconstruction ranged from C1 to C7 were performed on all subjects preoperatively and 1 week after surgery, using a GE Light Speed 64-slice VCT instrument (GE Healthcare, USA). CT parameters as follows: slice thickness, 0.625 mm; tube current, 220 mA; tube voltage, 120 kV; reconstruction interval, 0.625 mm; pitch, 0.925; gantry rotation time, 0.5 s; matrix, 512×512; FOV, 200×200 mm. Axical CT sections at each pedicle level were obtained for assessment to keep the same measurement position before and after surgery.

The CSA and SCD were assessed using the Picture Archiving and Communication System (PACS) (Carestream Vue; Carestream Health, Rochester, NY, USA), as shown in Figure 1, in which the CSA of the spinal canal was measured using an imaginary red line encircling the area with the intervertebral disc or the posterior edge of the vertebral body as the anterior boundary, the anterior edge of the vertebral arch as the posterior boundary, and the inner edge of the pedicle as the left and right borders. The SCD was measured using an imaginary yellow line connecting the Hridayesh midpoint of the posterior edge of the vertebral body to the top of the vertebral arch. Once the measurement area and distance were manually defined, the software automatically generated the measurement results. All the measurements (accuracy, 0.01 mm2 and 0.01 mm) were performed by 3 radiologists who were unaware of our study design; subsequently, these measurements were averaged for statistical analysis.

During the process of lamina opening, some vertebral laminas were inevitably worn off by the high-speed drill (Figure 2), and the sizes of lamina abrasion corresponding to the drill bit’s diameter (d). We set the mini-plate size to a, then the ELOE equals to the difference value between a and d.

STATISTICAL ANALYSIS:

The statistical analysis was conducted on SPSS software v24.0 (SPSS, Inc., Chicago, IL, USA). All variables were tested by one-sample Kolmogorov-Smirnov test and presented as the mean±SD at a significance level of P<0.05. To verify the effectiveness of the surgery, a paired t test was conducted to compare the preoperative and postoperative SCD/CSA. Linear and curvilinear regression models (first- and second-order polynomials) were used to investigate the correlation between the postoperative SCD/CSA increase and the ELOE. A higher R² value indicates a stronger correlation.

Results

GENERAL INFORMATION:

From March 2020 to December 2021, a total of 138 patients (76 males and 62 females) treated by UDCL in our institution were included, aged 38 to 83 (58.3±11.3) years, course of disease 4 to 48 (7.5±5.8) months, and the average duration of follow-up was 18.62±5.88 months. Pathologies of the patients included cervical spondylosis myelopathy (n=92) and ossification of the posterior longitudinal ligament (n=46). The presenting symptoms were: paresthesia in the extremities (n=92), intermittent gait incoordination (n=63), wide-based stance (n=52), radicular pain syndrome (n=43), and clumsiness of fine movements (n=26). All UDCL surgeries were successfully performed, the average operating time was 192.23±75.34 min, and blood loss was 182.36±52.37 ml.

ANALYSIS OF THE CERVICAL VERTEBRAE INVOLVED AND THE MINI-PLATE SIZE USED IN IHE SURGERY:

The cervical vertebrae concerned by UDCL were: C3–C7 (n=70, 50.72%), C3–C6 (n=36, 26.09%), C4–C7 (n=12, 8.70%), C4–C6 (n=9, 6.52%), C3–C5 (n=6, 4.35%), and C5–C7 (n=5, 3.62%), and all operations have involved C5. The sizes of the mini-plates used in the operation were: 12 mm (n=402, 66.78%), 14 mm (n=139, 23.09%), 10 mm (n=36, 5.98%), and 16 mm (n=25, 4.15%) (Figure 3). Among the 602 mini-plates used in surgeries, the 12-mm mini-plate was the most often used, while 16-mm plates were least used.

COMPARISON BETWEEN PREOPERATIVE AND POSTOPERATIVE SCD/CSA AND JOA SCORES:

The SCDs (Table 1) and CSAs (Table 2) were increased significantly after surgery in all cases (P<0.01). The JOA scores before and after surgery were 3.22±2.18 and 11.35±4.36, respectively, the improvement rate was 57.68±10.39%, and the JOA scores were significantly increased after surgery (P<0.01).

ANALYSIS OF THE RELATIONSHIP BETWEEN POSTOPERATIVE SCD/CSA AND ELOE:

A strong linear correlation between the ELOE and postoperative SCD increase was found by regression analysis (R2>0.939, P<0.01). The overall calculated postoperative SCD increase was approximately 0.519 times that of ELOE (Table 3).

In addition, the results of regression analysis indicated a strong linear correlation between the postoperative CSA increase and ELOE (CSA: R2>0.938, P<0.01). The overall calculated postoperative CSA increase was approximately 7.727 times that of ELOE (Table 4).

Discussion

After more than 40 years of development, the current trend of laminoplasty is developing toward the goal of muscle preservation and minimum tissue damage [14]. Given this, UDCL is superior to classic laminoplasty due to the biomechanical effect of the posterior musculature reconstruction, with a retained postoperative range of motion and loading response levels of the functional spinal units [3,8]. Several studies have confirmed that the use of the Centerpiece mini-plate can effectively avoid reclosure of the lamina after UDCL and provide adequate protection for the dural sac in the spinal canal, as well as avoid postoperative scar hyperplasia [1,15]. Finite element analysis showed better stability and more motion-sparing using the mini-plate in UDCL [8]. Either insufficient or excessive LOE can result in undesirable treatment outcomes, which is closely related to the mini-plate size [6,7]. Moreover, improper mini-plate used during surgery can also lead to poor biomechanical performance [16]. Therefore, selecting an optimal mini-plate size before surgery is rather important. However, due to the lack of a unified standard, most surgeons often select a mini-plate size based on clinical experience, which increases the possibility of poor treatment outcomes.

Previous studies have confirmed that the recovery of postoperative neurological function is associated with SCD and CSA [17,18]. In addition, for alleviating spinal cord compression, postoperative SCD increase appears to be more important than CSA increase [11]. Lubelski et al [19] considered that the probability of postoperative C5 palsy could decrease by 69% with every 1-mm increase in SCD. Since LOE determines postoperative SCD and CSA, the association between LOE and SCD/CSA has been investigated by numerous clinical studies. Itoh et al [20] suggested that the optimal SCD increase is 4.1 mm and the corresponding LOE is 8 mm. Another study suggested that the optimal SCD increase after UDCL is 4–5 mm [13]. Although the size of the mini-plate plays an important role in LOE, these 2 sizes are not exactly equal. It is worth noting that the above-mentioned related studies ignore the lamina abrasion extracted by drill during the operation, which may lead to unreliable results [1,15]. In addition, due to the irregular geometric structure of the cervical spinal canal, using SCD alone is still insufficient to predict and evaluate the treatment outcome, and postoperative CSA increase is also an important indicator [21]. Relevant research results are complex and limited to clinical application. The present study aimed to develop the concept of ELOE with consideration of the lamina abrasion extracted by the drill (Figure 2), in order to analyze the relationship between LOE and SCD/CSA more accurately.

All surgeries were successfully performed, as verified by a significant increase in SCD, CSA, and JOA scores after surgery. The results of regression analysis indicated a strong linear correlation between the ELOE and postoperative SCD increase, as well as postoperative CSA increase and ELOE. Although there was a linear correlation in each segment, for simplicity’s sake, the equation of C3–7 is recommended for use in clinical practice. According to the equation of C3–7, for each 1 mm increase of ELOE, the SCD would increase by about 0.5 mm. If a 12-mm mini-plate is used during the operation and the diameter of the drill is 2 mm, the ELOE is 10 mm, and the postoperative SCD increase is 5 mm. In the same way, the increase in postoperative CSA could also be estimated according to the size of mini-plate and the diameter of the drill, which provides an effective reference for the selection of intraoperative mini-plate. No standard regarding the most optimal postoperative SCD and CSA increase for UDCL has been proposed previously [9,10], and the finding of the present study only provide a prediction method based on mini-plate size and diameter of the drill, which should be applied flexibly in clinical practice.

It is also found that all operations in this study involved C5, and whether this is related to the occurrence of postoperative C5 palsy warrants further investigate. We performed a statistical analysis of the mini-plate sizes used during surgery, showing that 12-mm mini-plates were most often used (66.78%), while 10-mm and 16-mm mini-plates were least used (5.98% and 4.15%, respectively). Considering that insufficient or excessive LOE may lead to poor treatment outcomes, 10-mm and 16-mm mini-plates should be carefully selected in UDCL.

It is worth noting that there are some limitations in the present study. The lateral hinge location of the lamina is usually on the medial edge of the facet joint [1,6,14], but due to anatomical differences, even when the operations are conducted by a single surgical team, it is still difficult to identify the same hinge location. Besides, due to differences in quality of bone among different patients, the sizes of lamina abrasion extracted in UDCL could also be influenced; therefore, for patients with severe osteoporosis, careful consideration is needed when applying the results of this study. It should also be pointed out that, during the operation, the drill is expected to be perpendicular to the lamina, but this is commonly ignored in clinical practice, causing the lamina abrasion to be slightly larger than the diameter of the drill. To minimize this error, the surgical team should strictly adhere to operating standards to reduce lamina abrasion. Last but not the least, precise implantation of mini-plates is also important; if not, the actual LOE would be larger than the calculated value; therefore, skilled and careful manipulation is also required.

Conclusions

The present study developed the concept of ELOE considering the lamina abrasion in UDCL. A strong linear correlation was found between ELOE and the postoperative SCD/CSA increase. The present study provides a non-invasively tool for the optimal mini-plate selection prior to surgery, reducing the risk of adverse events such as mini-plate deformation or misplacement, and further improving the effectiveness and reliability of UDCL.