DOI: https://doi.org/10.21203/rs.3.rs-1724467/v1
The retrospective study was conducted to compare the efficacy of anterior cervical discectomy and fusion (ACDF) and anterior cervical corpectomy and fusion (ACCF) for localized ossification of the posterior longitudinal ligament (OPLL) by evaluating clinical and radiologic outcomes. We reviewed 151 patients to assess the effects of treatment for localized OPLL. Perioperative parameters, such as blood loss, operation time and complications, were recorded. Radiologic outcomes, such as the occupying ratio (OR), fusion status, cervical lordosis angle, segmental angle, segmental height, T1 slope, and C2–C7 sagittal vertical axis (SVA), were assessed. Clinical indices, such as the JOA scores and VAS scores, were investigated to compare the two surgical options. There were no significant differences in the JOA scores or VAS scores between the two groups (P > 0.05). The operation time, volume of blood loss and incidence of dysphagia were significantly less in the ACDF group than in the ACCF group(P < 0.05). In addition, cervical lordosis, segmental angle and segmental height were significantly different from their preoperative evaluations. Both groups had significantly improved T1 slopes. Moreover, C2–C7 SVA was significantly increased at the last follow-up. No adjacent segment degenerated in the ACDF group. The degeneration of the ACCF group was 4.1%. The incidence of CSF leaks was 7.8% in the ACDF group and 13.5% in the ACCF group. All the patients ultimately achieved successful fusion. Although both options achieved satisfactory primary clinical and radiographic efficacies, ACDF was associated with a shorter surgical procedure, less volume of intraoperative blood loss, better radiologic outcomes and lower incidence of dysphagia than ACCF.
Since its first reports in 1960 by Tsukimoto[1], ossification of the posterior longitudinal ligament (OPLL) is a common cause of cervical spondylosis. With the advancement of imaging technology, it has become increasingly common due to improvements in detection, especially in Asian countries. Although the reported prevalence of OPLL in Asians ranges from 1.9–4.3%[2], most patients with OPLL appear to be asymptomatic[3]. It is defined as an ectopic bone formation of the posterior longitudinal ligament[4]. It seems to be a persistent pathological development process. When an ossified ligament results in spinal stenosis, clinical symptoms and secondary cervical spondylosis can develop[5]. Operative treatment can block the natural process of OPLL to prevent further deterioration in compression of the spinal cord[6].
OPLL is morphologically classified into localized, segmental, continuous and mixed types[7]. Localized OPLL refers to an ectopic bone formation in the disc space or extending to the posterior border of the vertebral body[8]. Anterior surgery allows resection of the ossified mass to achieve direct decompression because the primary pathological mechanism of OPLL is spinal cord compression. ACDF and ACCF are two practical methods that are used for spinal decompression and cervical lordosis reconstruction for the treatment of severe spinal stenosis[9]. However, the subject of alternative surgical modalities is still controversial. No criteria have been established to decide the surgical procedure.
ACCF may be required if the osteophyte extends to the posterior fringe of the vertebral body. Therefore, most surgeons prefer anterior ACCF because it is easier to extirpate ossification. However, many complications frequently occurs in ACCF[2]. It has also been demonstrated that ACDF leads to lower perioperative complication rates than ACCF[10]. We can perform ACDF in patients with localized OPLL to avoid the removal of vertebral bodies. Smith, Robinson and Cloward described ACDF as a safe and efficient strategy for degenerative cervical spondylosis in 1958[11]. However, to date, few studies have compared the clinical efficacy of these two techniques in the treatment of localized OPLL, and little is known about the outcomes of ACDF for localized OPLL. So, in this retrospective study, we analyzed the clinical and radiographic outcomes to compare the efficacy of ACDF and ACCF.
This retrospective and comparative clinical study was from January 2018 to January 2021. We selected a total of 151 consecutive patients based on the timing of presentation who were without lost follow-up. They were subsequently divided into two groups based on surgical approaches applied (77, ACDF group; 74, ACCF group). The same physician recorded and evaluated the results of the follow-ups. All procedures performed in study were approved by the Institutional Ethics Committee of Soochow University. All experiments were performed in accordance with relevant named guidelines and regulations. The surgeries were performed by the same experienced physician.
The following were inclusion criteria: (1) refractory cervical radiculopathy with or without cervical myelopathy caused by localized OPLL; (2) findings showing localized OPLL by computed tomography(CT), and magnetic resonance imaging(MRI); (3) underwent anterior cervical surgery. Patients were excluded when (1) no symptoms caused by localized OPLL; (2) other types of OPLL (3) received other cervical spine surgery for tumors, fractures and so on; (4) anterior surgery combined with posterior surgery simultaneously.
We evaluated the clinical outcomes by JOA scores and VAS scores. Hirabayashi et al.[12] presented a method to calculate the recovery rate (RR): RR=(postoperative JOA score - preoperative JOA score)/(17 - preoperative JOA score)*100%. According to the Bazaz grading system (Table 1), we divided the severity of dysphagia. According to the Odom criteria (Table 2), we evaluated the surgical effects in each patient in the final follow-up.
Symptom severity | Liquid food | Solid food |
---|---|---|
None | None | None |
Mild | None | Rare |
Moderate | None or rare | Occasionally (only with specific food) |
Severe | None or rare | Frequent (majority of solids) |
Grade | Definition |
---|---|
Excellent | All preoperative symptoms relieved, able to carry out daily occupations without impairment |
Good | Minimum persistence of preoperative symptoms, able to carry out daily occupations without significant interference |
Fair | Relief of some preoperative symptoms, but whose physical activities were significantly limited |
Poor | Symptom and signs unchanged or worse |
Figure 1 shows the measurement of cervical spine parameters on the X-ray. (A)Cervical lordosis was measured by Cobb angle of C2-7 which formed between the inferior endplate of C2 and C7; (B)Segmental angle was the angle between the superior and inferior endplates of the fused vertebral body; (C) Segmental height was the distance between the midlines of the involved cranial and caudal vertebral bodies; (D) C2-7 sagittal vertical axis was the length from the postero-superior corner of C7 and the vertical line from the center of the C2 body; T1 slope was the angle between the upper endplate of T1 and the horizontal line. The CT scan can measure the occupying ratio (OR), which was determined by the ratio of the maximum anteroposterior diameter of the OPLL to the anteroposterior diameter of the spinal canal. Fusion was considered when there was no activity between the spinous processes, no radiolucent gap between the graft and endplate and continuous bridging of the trabecular bone at the graft and plate interface[13].
We analyze all statistics by performing SPSS version 26.0 (Chicago, IL, USA). Quantitative results are expressed as the mean and standard deviation. We used an independent t test for intergroup comparisons. We performed a paired t test before and after the operation. The chi-square test was used to compare the complications between the groups. When P values were less than 0.05, the results were recognized as significantly different.
151 patients successfully received ACDF and ACCF. The mean follow-up time was 23.3 ± 5.2 months, which ranged from 12 to 36 months. Table 3 shows detailed demographic characteristics. The mean age was 56.1 ± 8.1 years old of 77 patients in the ACDF group and that was 56.1 ± 9.8 years old of 74 patients in the ACCF group. The ossification levels in the ACDF group included C3-4(16 patients), C4-5(20 patients), C5-6(22 patients), C6-7(10 patients), C3-4 + C4-5(4 patients) and C4-5 + C5-6(5 patients). Another group included C3-4(16 patients), C4-5(18 patients), C5-6(24 patients), C6-7(10 patients), C3-4 + C4-5(3 patients) and C4-5 + C5-6(3 patients). No significant differences were found in age, sex, or ossification levels.
ACDF group | ACCF group | P value | |
---|---|---|---|
Sex(male/female) | 37/40 | 26/48 | 0.108 |
Age(years) | 56.1 ± 8.1 | 56.1 ± 9.8 | 0.988 |
Follow-up time(months) | 23.4 ± 5.4 | 23.3 ± 5.1 | 0.865 |
Ossification levels | 0.678 | ||
C3-4 | 16 | 16 | |
C4-5 | 20 | 18 | |
C5-6 | 22 | 24 | |
C6-7 | 10 | 10 | |
C3-4 + C4-5 | 4 | 3 | |
C4-5 + C5-6 | 5 | 3 | |
OR(%) | 40.6 ± 6.3 | 39.6 ± 10.1 | 0.194 |
Operative time(min) | |||
One-level | 82.6 ± 6.9 | 131.4 ± 24.5 | 0.000 |
Two-levels | 128.3 ± 11.3 | 176.2 ± 23.1 | 0.046 |
Blood loss(ml) | |||
One-level | 37.9 ± 15.4 | 129.0 ± 22.5 | 0.000 |
Two-levels | 58.3 ± 12.5 | 162.5 ± 26.6 | 0.000 |
ACDF anterior cervical discectomy and fusion | |||
ACCF anterior cervical corpectomy and fusion | |||
OR Occupation Rate |
For patients with single-space ossification, we performed single-level ACDF or single-level ACCF. For patients with dual-space ossification, we performed two-level ACDF or single-level ACCF. In the ACDF group, the mean operation time for one level was 82.6 ± 6.9 min. In the ACCF group, the mean operation time was 131.9 ± 24.5 min. The corresponding volumes of blood loss were 37.9 ± 15.4 and 129.0 ± 22.5 ml, respectively. The operation times for the two levels were 128.0 ± 11.3 and 176.2 ± 23.1 min, respectively. The corresponding blood loss was 58.3 ± 12.5 and 162.5 ± 26.6 ml, respectively. At either the single or two levels, ACDF significantly reduced the time of the surgical procedure and the volume of intraoperative blood loss when compared with ACCF (P < 0.05).
Table 4 shows that the mean JOA scores increased, and the corresponding VAS scores decreased in both groups. The JOA scores were 8.1 ± 1.3 increased to 15.8 ± 0.9 in the ACDF group and 8.1 ± 1.2 increased to 15.6 ± 1.2 in the ACCF group. The JOA recovery rates were calculated to be 58.9 ± 14.0 and 56.6 ± 17.6% postoperative in one month, respectively. At the final follow-up, they were 86.9 ± 11.2 and 83.6 ± 13.6%. The VAS scores were 7.6 ± 1.3 decreased to 1.5 ± 1.0 in the ACDF group and 7.6 ± 1.2 decreased to 1.5 ± 0.9 in the ACCF group. According to the Odom criteria, the percentages of patients with excellent and good clinical outcomes were 84.4% in the ACDF group and 83.8% in the ACCF group. An increase in JOA scores and a reduction in VAS scores were significant in both groups when compared with the initial measurement (P < 0.001). We found no significant differences between both groups at the baseline and each time point (P > 0.05).
ACDF group | ACCF group | P value | |
---|---|---|---|
JOA scores | |||
Preoperative | 8.1 ± 1.3 | 8.1 ± 1.2 | 0.992 |
Postoperative in 1 month | 13.4 ± 1.1# | 13.2 ± 1.4# | 0.319 |
Final follow-up | 15.8 ± 0.9# | 15.6 ± 1.2# | 0.092 |
RR(%) | |||
Postoperative in 1 month | 58.9 ± 14.0 | 56.6 ± 17.6 | 0.371 |
Final follow-up | 86.9 ± 11.2# | 83.6 ± 13.6# | 0.104 |
VAS scores | |||
Preoperative | 7.6 ± 1.3 | 7.6 ± 1.2 | 0.993 |
Postoperative in 1 month | 3.6 ± 1.2# | 3.6 ± 1.1# | 0.936 |
Final follow-up Clinical outcomes according to Odom criteria | 1.5 ± 1.0# Excellent: 30 Good: 35 Fair: 10 Bad: 2 | 1.5 ± 0.9# Excellent: 32 Good: 30 Fair: 9 Bad: 3 | 0.972 |
JOA scores Japanese Orthopedic Association scores | |||
RR Recovery Rate | |||
VAS scores Visual Analogue Scale scores | |||
# P < 0.05 comparing with preoperative value |
Figure 2 and Fig. 3 show representative images for one patient in each group. Table 5 lists a summary of the radiographic outcomes. The two groups showed a similar baseline before surgery (P > 0.05). The occupying rate in both groups were 40.6 ± 6.3% and 39.1 ± 10.1% which showed the similar severity of OPLL. In the ACDF group, the segmental height was 5.1 ± 1.1 mm preoperatively and 7.3 ± 0.8 mm in the final follow-up, the cervical lordosis was 16.4 ± 12.4° preoperative and 24.5 ± 9.2° in the final follow-up. In the ACCF group, the segmental height was 4.9 ± 0.9 mm preoperatively and 6.0 ± 0.8 mm in the final follow-up, the cervical lordosis was 13.8 ± 9.7° preoperative and 18.1 ± 10.5° in the final follow-up. The segmental height, the cervical lordosis angle, and the segmental angle increased in the ACDF group, which was significantly better than the corresponding parameters in the ACCF group (P < 0.05). The T1 slope improved from 20.2 ± 5.0° to 23.8 ± 4.9° in the ACDF group and from 18.9 ± 4.9° to 20.7 ± 4.8° in another group. ACDF showed more improvement than ACCF (P < 0.05). The C2–C7 SVA increased from 15.4 ± 7.1 mm to 18.6 ± 7.8 mm in the ACDF group and from 14.8 ± 9.6 mm to 18.2 ± 7.8 mm in the ACCF group at the last follow-up. We did not observe a difference (P > 0.05). In six months after the surgery, 66.2% and 40.5% patients in the two groups showed continuous bridging of the trabecular bone on the X-ray. In the final follow-up, solid fusion was successfully achieved in all patients. Figure 4 shows a representative case.
ACDF group | ACCF group | P value | |
---|---|---|---|
Segmental height (mm) | |||
Preoperative | 5.1 ± 1.1 | 4.9 ± 0.9 | 0.194 |
Postoperative in 1 month | 8.0 ± 1.0# | 6.7 ± 0.9# | 0.000 |
Final follow-up | 7.3 ± 0.8# | 6.0 ± 0.8# | 0.000 |
Cervical lordosis(°) | |||
Preoperative | 16.4 ± 12.4 | 13.8 ± 9.7 | 0.162 |
Postoperative in 1 month | 19.2 ± 11.3# | 15.8 ± 10.2# | 0.039 |
Final follow-up | 24.5 ± 9.2# | 18.1 ± 10.5# | 0.000 |
Segmental angle(°) | |||
Preoperative | 5.1 ± 5.1 | 3.8 ± 4.4 | 0.087 |
Postoperative in 1 month | 7.4 ± 4.7# | 5.4 ± 3.8# | 0.006 |
Final follow-up | 8.6 ± 5.0# | 5.2 ± 3.6# | 0.000 |
T1 slope(°) | |||
Preoperative | 20.2 ± 5.0 | 18.9 ± 4.9 | 0.110 |
Postoperative in 1 month | 22.9 ± 4.8# | 20.1 ± 4.5# | 0.000 |
Final follow-up | 23.8 ± 4.9# | 20.7 ± 4.8# | 0.000 |
C2-7 SVA(mm) | |||
Preoperative | 15.4 ± 7.1 | 14.8 ± 9.6 | 0.664 |
Postoperative in 1 month | 15.7 ± 4.8 | 15.6 ± 7.2 | 0.895 |
Final follow-up | 18.6 ± 7.8# | 18.2 ± 7.8# | 0.763 |
Fusion rate(%) | |||
Postoperative in 6 months | 51/77(66.2%) | 30/74(40.5%) | 0.002 |
Postoperative in 1 year | 71/77(92.2%) | 64/74(81.1%) | 0.254 |
Final follow-up | 77/77(100%) | 74/74(100%) | |
SVA Sagittal Vertical Axis | |||
# P < 0.05 comparing with preoperative value |
Table 6 presents a summary of the complications. After the surgery, 16 patients (20.8%) in the ACDF group and 30 patients (40.5%) in the ACCF group complained of dysphagia. One month postoperatively, eight patients and 12 patients disappeared, respectively. Finally, only four patients and 12 patients had no apparent relief. The difference in the occurrence of dysphagia was statistically significant at each follow-up between the two groups (P < 0.05). No ASD was observed in the ACDF group, while the degeneration of the ACCF group was 4.1%. CSFL occurred in three (7.8%) ACDF operations and in five (13.5%) ACCF operations. The incidence of ASD and CSFL between the two groups showed no significant difference (P > 0.05).
ACDF group | ACCF group | P value | |
---|---|---|---|
Dysphagia | |||
Postoperative | 16/77(20.8%) | 30/74(40.5%) | 0.008 |
Postoperative 1 month | 8/77(10.4%) | 18/74(24.3%) | 0.023 |
Final follow-up | 4/77(5.2%) | 12/74(16.2%) | 0.028 |
ASD | 0/77(0%) | 3/74 (4.1%) | 0.074 |
CSFL | 6/77(7.8%) | 10/74 (13.5%) | 0.254 |
ASD Adjacent Segment Degeneration | |||
CSFL Cerebrospinal Fluid Leak |
OPLL has a continuous pathology and may progress more aggressively when secondary to myelopathy. OPLL can be removed by anterior approaches. Although doing so is more technically demanding, it allows direct decompression, which may be related to better prognosis and less progression[14]. Localized OPLL is usually asymptomatic at the very beginning. Patients’ symptoms are aggravated by disc herniation at the same intervertebral space level or due to neurological symptoms caused by disc injury as a result of recent trauma. For this reason, ACDF can achieve the same efficacy as ACCF by resecting the herniated disc and OPLL at the intervertebral space level, with much less surgical injury than ACCF.
In this study, we found that the volume of blood loss and duration of surgical procedure were significantly less in patients undergoing ACDF than in patients undergoing ACCF. The majority of previous studies were in accordance with our results[15]. This difference is because ACCF requires resecting the corresponding vertebral body and the adjacent disc, which is more traumatic. However, Lin et al.[16] found that the duration of the ACDF surgical procedure was significantly longer than that of the ACCF procedure, which was contrary to our study. He believed that performing ACDF to remove osteophytes from the intervertebral space is time-consuming and more technically challenging. The reason why many scholars perform ACCF is that the difficulty of ACDF is much greater than that of ACCF. The limited surgical exposure and the adhesion of the OPLL to the dura make the operation more difficult. Therefore, ACDF was performed only when the preoperative CT scan showed that the OPLL did not extend to the entire intervertebral space.
ACCF and ACDF not only relieves compression in the spinal cord but also reconstructs the physiological parameters of the cervical spine through careful distraction of the intervertebral space[17]. Burkardt et al.[18] showed that ACDF had better maintenance of segmental height and greater improvements in cervical lordosis. Our study corresponded with it. Compared with patients who underwent ACCF, the postoperative segmental height was significantly improved in those who underwent ACDF. However, segmental height dropped slightly in both groups when they came to review and the ACCF group was more than the ACDF group. The reduced height of the treated segment may be due to postoperative subsidence of implant settling and then migrating to the endplates. Park et al.[19] also expounded subsidence of the treated intervertebral space height at six weeks after surgery. In addition, the thickness of the removed endplate also had a significant effect on the subsidence of the cage after fusion surgery[20]. The main causes of abnormalities in patients who undergo ACCF may be related to excessive damage to the vertebral endplate and the rigid effect of the titanium cage. The contact surface of the titanium cage is sharper. Under the action of stress, it is more likely to protrude to the cancellous bone to form subsidence. However, the cage used in ACDF has a relatively large effective contact surface, which disperses the stress of adjacent vertebrae and is more advantageous in the maintenance of intervertebral height.
Biomechanically, we believe that multiple points of distraction during ACDF can more effectively correct the cervical curvature. Clinical studies have directly linked postoperative cervical kyphosis to greater neurological deficits[21]. In our study, the C2-7 Cobb angle and the segmental angle in ACDF significantly showed greater improvement than that in ACCF at each follow-up. Therefore, we speculate that the maintenance of lordosis after ACDF may be better than that after ACCF from a long-term perspective[22]. Other main radiologic factors associated with it were T1 slope and C2–C7 SVA. Zhang et al.[23] suggested that the sagittal balance of ACDF was better than that of ACCF, which was correlated with fewer axial symptoms. From our study, we found that the T1 slope had a greater improvement in the ACDF group than in the ACCF group. Thus, two approaches corrected the balance by ameliorating the T1 slope. We found that no significant difference was observed in C2–C7 SVA in our research, and the values were at a normal level. MCAVINEY et al.[24] found that the gravity center of the head will shift forward when the C2-7 SVA is greater than 40 mm, which will straighten cervical curvature and affect horizontal vision. This finding indicated that two surgical procedures caused little damage to the neck muscle, could prevent the cervical gravity center from moving forward, better maintained cervical sagittal balance and enhanced the postoperative quality of life of patients.
The patients had similar chief complaint symptoms before surgery. Both clinical outcomes showed no significant difference, which was similar to previous studies[15]. Moreover, postoperative clinical outcomes improved significantly in both groups. This finding indicated that patients improved after surgery, and their condition gradually improved at each follow-up. Thus, the effect of the anterior approach is acceptable. Statistics show that reports of improvement in neurological function is approximately between 60% and 70%[25]. The JOA recovery rate in our study improved greatly. During the follow-up, almost all patients were satisfied with the curative effect. They obviously felt that the neck pain disappeared, and the upper limb numbness was relieved. Wang et al.[26] reported that either ACCF or ACDF had a highly successful improvement of clinical outcomes and that they are good solutions.
The success of cervical surgery often depends directly on the progress of fusion[27]. From a biomechanical point of view, additional external or internal support should be employed to prevent excessive movement of cervical cages[28]. Qiu et al.[29] reported that corpectomy and discectomy with plate fixation and autograft fusion had similar fusion rates. Several scholars have shown that ACCF has relatively good fusion rates[30]. Pseudarthrosis may occur in multi-segmented ACDF if fusion surfaces are increased[31]. This might be the reason for the fusion failure. However, in our study, the six-month fusion rates of patients who underwent ACDF and those who underwent ACCF were 66.2% and 40.5%, respectively. This is statistically significant. This result indicated that ACDF might promote earlier solid fusion in the six months. At one year, the difference disappeared. In the final follow-up, patients achieved successful solid fusion in each group.
In anterior cervical spine surgery, chronic dysphagia is one of the most common postoperative complications, but the mechanism remains unknown. Esophageal injury, anterior cervical soft tissue edema and postoperative hematoma might contribute to dysphagia[32]. In ACDF, many methods can be used for final internal fixation. We preferred to employ ROI-C, which is a novel zero-profile anchored spacer. Liu et al.[33] conducted a comparative study in which ROI-C was linked with a simpler surgery and a lower incidence of complications when compared to PEEK cages with anterior plates. In the ACCF group, we still employed the plate traditionally. Sixteen (20.8%) patients complained of mild dysphagia postoperatively in the ACDF group and thirty (45.9%) in the ACCF group after surgery. Some of them might be due to recurrent laryngeal nerve injury, and they recovered after we used methylprednisolone. We speculate that ACDF requires a smaller incision with less exposure and can avoid mechanical stimulation of the associated structures. At one month postoperatively, we still found a lower incidence of dysphagia in patients who underwent ACDF (10.4%) than in those who underwent ACCF (24.3%). We thought that the plate used in ACCF made the difference and that the presence of a plate caused anterior cervical soft tissue edema. In ACDF, We implanted a special cage into the intervertebral space and insert two anchoring clips to fix the cage into the adjacent vertebra[34]. This method could prevent the implant from making contact with the anterior cervical soft tissue. Through our follow-up of patients, only one patient in the ACDF group and five patients in the ACDF group had no apparent relief finally. Thus, we conclude that ACDF that uses ROI-C is great at ensuring milder anterior soft tissue injury, which results in a lower incidence of dysphagia. Another long-term postoperative complication is ASD, because cervical spine fusion surgery immobilizes the motion segment at the expense of its range of motion. Fewer remaining segments are considered to provide more motion, which accelerates disc degeneration, chronic osteophyte formation and new adjacent segment disease[34]. Another factor is that the edge of the plate approaches the disc[35]. However, the results showed in our current study that no case of ASD was observed in the ACDF group, and the incidence was 4.1% in the ACCF group. Fortunately, the two patients showed only changes in imaging data and no definite clinical symptoms or mild symptoms, so no repeated operations were needed. The low incidence of ASD in this study may be due to the short observation period. Nevertheless, we speculate that the incidence of ASD will increase in those who undergo ACCF in the future because of the inevitable misposition of the plate.
Cerebrospinal fluid leak (CSFL) after cervical spine surgery can be troublesome, as it can cause meningitis, spinocutaneous fistula, or pseudomeningocele. Lee et al.[36] reports that dural tears are relatively common, as they occur in approximately 10.5% of ACCF procedures. The crucial point of our operation is that the herniated disc is completely removed first. Then, we can incise the normal posterior longitudinal ligament so that we can, in turn, expose the dura and remove the OPLL with a rongeur. If the posterior longitudinal ligament that is adhered to the dura cannot be removed, then we dissect the OPLL and allow it to float on the surface of the dura to relieve compression. When a rongeur was used to resect the localized OPLL, it was much more likely to break the dura if the OPLL was adhered to the dura. In our study, CSFL occurred in six patients in the ACDF group and in ten patients in the ACCF group. Although the incidence was not statistically significant, we considered that resection of more structures during ACCF is more likely to cause dural tears. Various techniques have been used to manage dural tears and the consequent CSFL after surgery. Gelatin sponges were successfully used to repair dural tears intraoperatively. These patients completely recovered after 3–5 days of bed rest and management with ceftriaxone, and their wounds healed normally. There was one patient in the ACCF group in which lumbar cistern drainage postoperatively was employed to solve the problem because the size of the dural defect was large. No patient developed further spinal or intracranial infection, cutaneous fistula or secondary surgery.
This study has some limitations. As a retrospective study, a randomized controlled trial study should be carried out, and a large sample is necessary. The current findings cannot be interpreted as long-term results. Therefore, we require further observation.
Both ACDF and ACCF achieved satisfactory primary clinical efficacies and fusion rates for patients with localized OPLL. However, ACDF was associated with a shorter surgical procedure, less volume of blood loss, better radiologic outcomes, and lower incidence of dysphagia than ACCF. In most cases, when both surgical approaches are an option, ACDF is a worthwhile alternative to ACCF as a surgical treatment for localized OPLL.
Author contributions Chen Tangyiheng, Wang yujie and Zhou hong wrote the main manuscript text. Lin cheng and Zhao Xingyi collected the data. Wang Genlin, Li xuefeng, Chu Genglei, Jiang Weimin and Liu yijie gave the suggestion for modification. All authors reviewed the manuscript.
Conflicts of interest The authors declare no potential confict of interest.
Funding The study is supported by the Youth Fund for Medical and Health Basic Research of Suzhou(SYS2020096)
Ethics declarations All procedures performed in study were approved by the Institutional Ethics Committee of Soochow University. All experiments were performed in accordance with relevant named guidelines and regulations.
Consent to participate Informed consent was obtained from all individual participants included in the study.
Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.