Heterotopic ossification of the posterior longitudinal ligament (OPLL) leads to narrowing of the spinal canal. In the Japanese population, extensive ossification often develops in the spinal ligaments, including the posterior longitudinal ligament.1 OPLL consequently represents one of the most common causes of cervical myelopathy in Japan, along with disc herniation and spondylosis.1 Because conservative treatment is usually ineffective for severe myelopathy caused by OPLL, surgical treatment is chosen in most cases. Decompressive surgical procedures for OPLL-related cervical myelopathy can be divided into those using an anterior approach and those using a posterior approach. Several reports have analyzed the surgical outcomes of anterior and posterior surgeries; however, evaluation of the surgical procedures differs between institutions.2–13 Furthermore, guidelines for selecting surgical procedures for cervical OPLL have not yet been fully established.
Since 1968, we have performed surgery using anterior or posterior approaches for patients with cervical myelopathy due to OPLL.6 Between 1968 and 1991, we have twice modified the procedures of our anterior and posterior surgeries. During the first phase, from 1968 to 1980, we performed anterior decompression with spinal fusion (ASF) and extirpation of OPLL within 3 intervertebral disc levels below C4 as the anterior procedure, and laminectomy for multilevel lesions over 3 intervertebral disc levels as the posterior procedure. During the second phase, from 1980 to 1986, we modified our anterior procedure by extending ASF up to 3 intervertebral disc levels below C3, and replaced laminectomy with laminoplasty as our posterior procedure. During the third phase, from 1986 to 1991, we further modified the anterior procedure by using a cervical distractor and intraoperative ultrasonography, and modified the posterior procedure by incorporating domelike enlargement of the C2 lamina and extirpation of the C1 posterior arch when necessary. In cases with anterior surgery, the first and second modifications of surgical techniques have improved neurologic recovery, respectively. Similar improvement was also observed in cases with posterior surgery. Our recent follow-up studies of cases during our third phase found that the surgical outcome of ASF was superior to that of laminoplasty; the mean recovery rate after ASF was 71.4% and that after laminoplasty was 61.4%, indicating that complete decompression of the spinal cord anteriorly may result in better neurologic recovery than indirect decompression posteriorly.6
Starting in 1993, we presented the recent surgical outcome and the standardized informed consent of both anterior and posterior surgeries to all cervical OPLL patients who were scheduled to undergo surgery at our institute. We selected the surgical approach in accordance with the patients' choice between anterior and posterior surgeries. In the present study, we compared the results of anterior and posterior surgeries performed after 1993, and we analyzed factors related to poor surgical outcomes.
MATERIALS AND METHODS
From June 1993 to July 2002, 66 patients with cervical myelopathy due to OPLL underwent surgical treatment (21 anterior, 45 posterior) at our institute. Of these 66 patients, 7 patients (2 anterior, 5 posterior) were excluded from this study because they had undergone follow-up evaluation of less than 1 year duration. The study group thus comprised the remaining 19 anterior surgery patients (A-group) and 40 posterior surgery patients (P-group) with 1 year or longer follow-up after surgery, a total of 59 patients (44 males and 15 females). Their mean age at surgery was 59.4 years, ranging from 38 to 82 years (Table 1).
Selection Between Anterior and Posterior Surgery
Before surgery, we provided standardized informed consent to all cervical OPLL patients who were scheduled to have surgery in our institute. We explained our recent surgical results of both ASF and laminoplasty to the patients, including neurologic recovery, surgery-related complications, and postoperative neck immobilization with a cervical orthosis. We selected the surgical approach in accordance with the patients' choice between anterior and posterior surgery.
We explained the relative advantages of anterior surgery (ASF) as follows: (1) excision of the ossified mass enables complete decompression of the spinal cord; (2) ASF creates a solid spinal fusion that can relieve pressure on the injured spinal cord. We also discussed the following disadvantages of ASF compared with laminoplasty: (1) more skillful techniques are required for ASF than for laminoplasty; (2) harvesting the graft bone from the iliac crest or fibula entails a risk of additional complications, such as donor site pain; (3) ASF requires a longer postoperative immobilization of the neck with a cervical orthosis; and (4) ASF is principally not applicable to the lesion above C2 and below T3.6,7,10,12
We also explained the relative advantages of posterior surgery (laminoplasty) as follows: (1) the surgical technique is less difficult for laminoplasty than for ASF; (2) the duration of hospitalization and postoperative immobilization of the neck with an orthosis is shorter for laminoplasty than for ASF. We also discussed the following disadvantages of laminoplasty compared with ASF: (1) decompression of the spinal cord after laminoplasty occurs by a posterior shift of the cord, that is, by so-called indirect decompression. If the posterior shift is not enough, anterior compression of the cord by the ossified mass may persist, leading to diminished recovery from myelopathy; (2) the frequency of postoperative neck pain due to injury to the nuchal muscles is generally higher after laminoplasty than after ASF.2,6,8,11,13
Our anterior surgery procedure (ASF) has been described in detail previously.6 The area of decompression and spinal fusion varies from a single intervertebral disc level to 4 intervertebral disc levels. We normally perform complete excision of the ossified mass, but when cerebrospinal fluid leakage and/or massive bleeding occurs during excision of the OPLL, we intraoperatively alter the procedure to the anterior floating method, in which the OPLL is incompletely removed, leaving behind a thin layer of OPLL on the reexpanded dura.7,10 Spinal fusion is performed with an autogenious iliac crest or fibula bone graft. Internal fixation devices such as a plate and screw system normally are not used.
Our posterior surgery procedure is cervical enlargement laminoplasty, normally consisting of a C3-C7 en-block laminoplasty (Itoh's method).3 Depending on the degree of cord compression, we sometimes add C2 dome laminotomy, resection of C1 posterior arch, and/or laminoplasty of T1.
Postoperatively, patients are allowed to sit up wearing a cervical orthosis within the first week after surgery. After an ASF of 1to 3 intervertebral disc levels, patients wear a molded cervicothoracic orthosis for 2 months and then a soft cervical collar for 1 month. After an ASF of four intervertebral levels (3-level corpectomy), patients are cared for in the intensive care unit with intratracheal tubing for the first 48 hours after surgery, and they wear a Halo-vest for at least 2 months. After laminoplasty, patients wear a molded cervicothoracic orthosis or a soft cervical collar for 1 to 3 months after surgery.
The Japanese Orthopedic Association (JOA) scoring system was used to evaluate the severity of cervical myelopathy (Table 2). The JOA scores before surgery and 1 year after surgery were evaluated, and the recovery rate was calculated. We defined the surgical outcome as follows: excellent (80%≤recovery rate), good (40%≤recovery rate <80%), fair (10%≤recovery rate <40%), unchanged (0%≤recovery rate 10%), or worsened (recovery rate <0%).
Extension and flexion radiographs were assessed before surgery and 1 year after surgery, and were used to measure the range of intervertebral mobility at the level of maximum spinal cord compression (segmental range of motion) (Fig. 1). The cervical lordotic angle (C2-C7 angle) was measured on lateral radiographs before surgery and 1 year after surgery (Fig. 1), and the change of cervical lordosis was calculated as follows: ΔC2-C7 lordotic angle=C2-C7 angle after surgery-C2-C7 angle before surgery. All patients had computed tomography before surgery, and the occupation ratio of the ossified mass at the most stenotic level of the spinal canal was defined as follows: OPLL occupation ratio=(thickness of OPLL/anteroposterior diameter of the bony spinal canal)×100.
The Mann-Whitney U test was used for unpaired data. P values <0.05 were considered significant. Results are presented as the mean±standard deviation of the mean.
Anterior Surgery Versus Posterior Surgery
Preoperative clinical data for the A-group and the P-group are summarized in Table 1. The mean age at surgery was 51.8 years for the A-group and 62.6 years for the P-group, indicating that at the time of surgery, the laminoplasty patients were significantly older than the ASF patients (P<0.01). The duration of symptoms (the period from the onset of myelopathy until surgery) was significantly longer for the laminoplasty patients: 29.1 months for the A-group vs. 50.0 months for the P-group (P<0.05). The mean number of ossified vertebra was 2.5 in the A-group and 4.1 in the P-group (P<0.01). The OPLL mean occupation ratio in the spinal canal did not significantly differ between the 2 groups: 56.0% for the A-group versus 55.9% for the P-group.
The clinical results of surgery for the A-group and the P-group are shown in Table 3. The mean preoperative JOA score was 8.3 points in the A-group and 8.6 in the P-group, a difference that was not significant (P=0.568). The mean JOA score 1 year after surgery was 14.2 points in the A-group and 13.0 points in the P-group, again a difference that was not significant (P=0.087). However, the mean recovery rate was significantly higher (P<0.05) in the A-group (68.4%) than in the P-group (52.5%), indicating that neurologic recovery was better after ASF than after laminoplasty. Fifteen patients had a recovery rate less than 40%: 2 in the A-group and 13 in the P-group. One laminoplasty patient and none of the ASF patients developed postoperative aggravation of their neurologic status.
Poor Surgical Outcome After Laminoplasty
The 40 P-group subjects were divided into 2 subgroups: a good recovery group comprising patients whose recovery rate was 40% or higher (n=27) and a poor recovery group comprising patients whose recovery rate was less than 40% (n=13). Possible factors affecting poor surgical outcome after laminoplasty are shown in Table 4. No significant differences were seen in the preoperative JOA score between the good and poor outcome groups. The age at surgery was significantly higher in the poor outcome group (mean, 68.0 y) than in the good outcome group (mean, 59.9 y) (P<0.05). The duration of symptoms was significantly longer in the poor outcome group (mean, 61.6 mo) than in the good outcome group (mean, 39.9 mo). No significant differences between the subgroups were observed in the OPLL occupation ratio. The change of the cervical lordotic angle after laminoplasty (ΔC2-C7 lordotic angle) was −1.44 degrees in the good outcome group and −4.85 degrees in the poor outcome group (P<0.05), indicating that the decrease of cervical lordosis after laminoplasty caused poor neurologic recovery. The mean range of motion at intervertebral disc level of maximum cord compression was 6.9 degrees in the good recovery group and 10 degrees in the poor outcome group before surgery (P<0.05), and 3.1 degrees in the good recovery group and 7.2 degrees in the poor outcome group after surgery (P<0.05), indicating that preoperative and postoperative segmental mobility was significantly greater in patients who had a poorer neurologic recovery.
A 68-year-old man presented with bilateral hand clumsiness and a spastic gait. His preoperative JOA score was 6.5 points. Lateral radiographs, computed tomography, and magnetic resonance images demonstrated a C2-C7 OPLL associated with severe compression of the spinal cord anteriorly at C3/4 (Figs. 2A–C). Extension and flexion radiographs showed that the range of motion at C3/4 was 8 degrees, indicating evident intervertebral mobility at the maximum cord compression level (Figs. 2D, E). Posterior surgery was selected, and a C3-C7 laminoplasty was performed. A postoperative radiograph 1 year after surgery demonstrated progression of cervical kyphosis (Fig. 2F), and magnetic resonance images showed persistent anterior impingement of the spinal cord by OPLL (Figs. 2G, H). Postoperatively, the patient had a poor neurologic recovery: his JOA score was 9 points 1 year after surgery, yielding a recovery rate of 21.7%. The patient was not satisfied with his degree of recovery from myelopathy.
Tani et al12 comparatively analyzed the clinical results of 26 patients who underwent microscopic anterior decompression and spinal fusion (n=14) or laminoplasty (n=12) for extensive cervical OPLL with an occupation ratio in the spinal canal exceeding 50%. They found that the average recovery rate was significantly higher after anterior surgery than after posterior surgery. No neurologic deterioration occurred after anterior surgery, whereas postoperative neurologic deterioration occurred in 4 patients after posterior surgery. The authors suggested that the following factors affected neurologic deterioration after laminoplasty: (1) decrease in the lordosis of the cervical spine; (2) tethering of nerve roots due to insufficient width for decompression of the spinal cord; (3) inappropriate positioning of the neck during surgery; (4) direct intraoperative damage to the spinal cord; and (5) ischemia of the spinal cord during surgery.
The present study yielded data consistent with those of Tani et al,12 demonstrating that the surgical outcome of anterior surgery for cervical OPLL was superior to the surgical outcome of posterior surgery. The present results also showed that the decrease of cervical lordosis after laminoplasty caused poor surgical outcome; being principally consistent with the description of Tani et al.12 In addition, the results demonstrated that elderly patients treated with laminoplasty were especially prone to a poor surgical outcome. Despite the preoperative information they received showing the good surgical outcome associated with anterior surgery, many elderly patients selected posterior surgery. Although they could understand the likelihood of a better neurologic recovery after anterior surgery, these elderly patients judged that the postoperative course of anterior surgery would be too difficult to tolerate.
In the present study, surgical approaches were determined in accordance with the patients' choice. Thus, the study groups were not truly randomized. Despite such limitation of the study design, the present data provided evidence that substantial intervertebral disc mobility at the maximum cord compression level could be a factor causing poor surgical outcome after laminoplasty. The study data demonstrated that the poor outcome group after laminoplasty had larger segmental mobility of vertebrae before surgery and after surgery. We suggest that laminoplasty in patients with massive OPLL may cause insufficient posterior shift of the spinal cord, resulting in persistent anterior impingement of the spinal cord by OPLL. If substantial segmental mobility still remains after surgery in such cases, it is possible that damage to the injured spinal cord progresses.
Based on the present results, we recommend that ASF should be the first choice of treatment for patients with massive OPLL and evident intervertebral disc mobility at the cord compression level. When laminoplasty is selected for such cases, the addition of posterior instrumented fusion would be desirable for stabilizing the spine and decreasing the damage to the cord.
In previous studies, we analyzed factors affecting poor surgical outcomes for thoracic myelopathy due to OPLL. We examined 3 surgical procedures: (1) extirpation of OPLL; (2) posterior decompression alone; and (3) posterior decompression with instrumented fusion. With respect to outcomes, 3 of the 16 patients who underwent posterior decompression alone developed paralysis after surgery. Two of these 3 postoperative paralysis patients fully recovered after they underwent posterior fixation with instrumentation.14,15 None of the patients who initially underwent posterior decompression and instrumented fusion for thoracic OPLL developed postoperative paralysis, and all of these patients experienced neurologic recovery despite anterior impingement of OPLL on the spinal cord.16 These findings suggested that the addition of instrumented fixation reduced the mobility of the spine and prevented damage to the spinal cord. Because most cases of cervical OPLL are associated with a lordotic alignment of the cervical spine, the surgical approach to cervical OPLL will usually differ from the approach to thoracic OPLL. In cervical OPLL patients who have massive OPLL and evident intervertebral disc mobility at the cord compression level, however, our experience with the treatment of thoracic OPLL indicates the potential use of posterior instrumented fusion. In 2003, we began to employ posterior decompression with instrumented fusion as a surgical procedure for elderly patients with cervical OPLL, and thus far have obtained better surgical outcomes from this procedure than from laminoplasty (to be published elsewhere).
1. Tsuyama N. Ossification of the posterior longitudinal ligament of the spine. Clin Orthop. 1984;184:71–84.
2. Hirabayashi K, Miyakawa J, Satomi K, et al. Operative results and postoperative progression of ossification among patients with ossification of cervical posterior longitudinal ligament. Spine. 1981;6:354–364.
3. Itoh T, Tsuji H. Technical improvements and results of laminoplasty
for compressive myelopathy in the cervical spine. Spine. 1985;10:729–736.
4. Wan-Chun C, Chen-Nen C, Tai-Ngar L, et al. Surgical treatment for ossification of the posterior longitudinal ligament of the cervical spine. Surg Neurol. 1994;41:90–97.
5. Kawano H, Handa Y, Ishii H, et al. Surgical treatment for ossification of the posterior longitudinal ligament of the cervical spine. J Spinal Disord. 1995;8:145–150.
6. Goto S, Kita T. Long-term follow-up evaluation of surgery for ossification of the posterior longitudinal ligament. Spine. 1995;20:2247–2256.
7. Yamaura I, Kurosa Y, Matsuoka T, et al. Anterior floating method for cervical myelopathy caused by ossification of the posterior longitudinal ligament. Clin Orthop. 1999;359:27–34.
8. Sodeyama T, Goto S, Mochizuki M, et al. Effect of decompression enlargement laminoplasty
for posterior shifting of the spinal cord. Spine. 1999;24:1527–1531.
9. Chiba K, Toyama Y, Watanabe M, et al. Impact of longitudinal distance of the cervical spine on the results of expansive open-door laminoplasty
. Spine. 2000;25:2893–2898.
10. Matsuoka T, Yamaura I, Kurosa Y, et al. Long-term results of the anterior floating method for cervical myelopathy caused by ossification of the posterior longitudinal ligament. Spine. 2001;26:241–248.
11. Iwasaki M, Kawaguchi Y, Kimura T, et al. Long-term results of expansive laminoplasty
for ossification of the posterior longitudinal ligament of the cervical spine: more than 10 years follow up. J Neurosurg. 2002;96:180–189.
12. Tani T, Ushida T, Ishida K, et al. Relative safety of anterior microsurgical decompression versus laminoplasty
for cervical myelopathy with a massive ossified posterior longitudinal ligament. Spine. 2002;27:2491–2498.
13. Ogawa Y, Toyama Y, Chiba K, et al. Long-term results of expansive open-door laminoplasty
for ossification of the posterior longitudinal ligament of the cervical spine. J Neurosurg Spine. 2004;1:168–174.
14. Yamazaki M, Okawa A, Koda M, et al. Transient paraparesis after laminectomy for thoracic myelopathy due to ossification of the posterior longitudinal ligament: a case report. Spine. 2005;30:E343–E346.
15. Yamazaki M, Koda M, Okawa A, et al. Transient paraparesis after laminectomy for thoracic ossification of the posterior longitudinal ligament and ossification of the ligamentum flavum. Spinal Cord. 2006;44:130–134.
16. Yamazaki M, Mochizuki M, Ikeda Y, et al. Clinical results of surgery for thoracic myelopathy due to ossification of the posterior longitudinal ligament: operative indication of posterior decompression with instrumented fusion. Spine. 2006;31:1452–1460.