Radiculopathy of C-5 after anterior decompression for cervical myelopathy

Transcription

1 J Neurosurg Spine 3: , 2005 Radiculopathy of C-5 after anterior decompression for cervical myelopathy MINORU IKENAGA, M.D., PH.D., JITSUHIKO SHIKATA M.D., PH.D., AND CHIAKI TANAKA M.D., PH.D. Department of Orthopedic Surgery, Kyoto City Hospital; and Department of Orthopedic Surgery, Yawata Central Hospital, Kyoto, Japan Object. The authors conducted a study to examine the incidence and causes of postoperative C-5 radiculopathy, and they suggest preventive methods for C-5 palsy after anterior corpectomy and fusion. Methods. The authors included in the study 18 patients with postoperative C-5 radiculopathy from 563 patients who underwent anterior decompression and fusion for cervical myelopathy. There were 10 cases of ossification of the posterior longitudinal ligament (OPLL) and eight cases of cervical spondylotic myelopathy (CSM). All patients received conservative treatment. Posttreatment full recovery was present in eight patients, and Grade 3/5 strength was documented in six in whom some weakness remained. Radiographic evaluation revealed that the C3 4 and C4 5 cord compression was significantly more severe in patients with paralysis than in those without paralysis. The incidence of paralysis was higher in patients with OPLL than in those with CSM (chi-square test, p = 0.03). The incidence of paralysis increased in parallel with the number of fusion levels (correlation coefficient r = 0.94). Multivariate analysis revealed that the final manual muscle testing (MMT) value was closely related to the preoperative MMT value (computed t value 4.17; p 0.01) and preoperative Japanese Orthopaedic Association (JOA) score for cervical myelopathty (computed t value, 2.75; p 0.05). Conclusions. Preexisting severe stenosis at C3 4 or C4 5 in patients with OPLL is a risk factor for paralysis. Preoperative muscle weakness and a low JOA score are factors predictive of poor recovery. KEY WORDS anterior fusion cervical spine palsy C-5 palsy cervical myelopathy F OR the treatment of cervical myelopathy, anterior or posterior decompression has been advocated, and both procedures have been reported to yield generally good results. 1,3,4,6,7,9,23,24 Radiculopathy of C-5, however, has been reported as a major complication of both procedures. The development of radiculopathy after decompressive laminoplasty has been intensively studied, 2,5,19,20,21,25 whereas to date there are only a few studies of this complication after anterior decompression. 8,10,12 15,22 Radiculopathy of C-5 after decompression and multilevel corpectomy was noted first in the English literature by Shinomiya, et al. 15 They suggested that it was due to an axial rotation deformity of the cervical spine. They later also reported that the anatomical features of the C-5 nerve root may be a cause of the palsy. 16 Yonenobu, et al., 25 discussed possible causes of postdecompression radiculopathy: C-5 nerve root tethering, the anatomical morphology Abbreviations used in this paper: AP = anteroposterior; CSM = cervical spondylotic myelopathy; CT = computerized tomography; JOA = Japanese Orthopaedic Association; MMT = manual muscle test; MR = magnetic resonance; OPLL = ossification of the posterior longitudinal ligament. 210 of the C-5 root, and the single-nerve innervation to the deltoid muscle. In a 1995 study of a series of anterior corpectomies, Saunders 13 stated that a decompression width greater than 15 mm might increase the incidence of radiculopathy. In our clinic, anterior decompression procedures have been preferred for cervical myelopathy since During a 13-year period ( ), the total number of anterior decompression procedures was 563, representing the largest series yet reported. The purpose of this study was to analyze the incidence and characteristics of C-5 radiculopathy of after anterior decompression and fusion. Clinical Material and Methods Five hundred sixty-three patients with cervical myelopathy underwent anterior decompression and fusion in our hospital from 1989 to There were 353 male and 210 female patients who ranged in age from 11 to 86 years. There were 137 cases of OPLL, 287 cases of CSM, 79 cases of disc herniation, and 60 cases involving other lesions (Table 1). We performed a one-level fusion in 64 patients, two-level fusion in 137, three-level fusion

2 Radiculopathy of C-5 after anterior cervical decompression TABLE 1 Cause of radiculopathy in the overall patient population* Origin No. of Cases cervical spondylosis 328 OPLL 137 disc herniation 79 tumor 9 trauma 8 RA 2 total 563 * RA = rheumatoid arthritis. in 154, four-level fusion in 189, five-level fusion in 17, and six-level fusion in two cases (Table 2). The mean number of fusion levels was 2.9. In the patients in whom the fusion involved more than four disc levels, a plate was not placed; instead, halo vest fixation was used for 6 weeks and Philadelphia collar fixation for the next 6 weeks thereafter. In patients in whom the fusion incorporated one to three disc levels, a plate was implanted for internal fixation, and Philadelphia collar fixation was subsequently instituted for 3 months. Eighteen patients (13 male and five female patients) suffered postoperative motor dominant paralysis in the upper extremities (C-5 radiculopathy) and were included in the study (Table 2). The medical records of these 18 patients were intensively reviewed, and patient age, sex, laterality of the lesion, time of onset, and muscle power were recorded (Table 3). Manual muscle testing (Grades 0 5/5) was performed sequentially to determine recovery after the onset of paralysis. Clinical results were assessed using the JOA tool for measuring cervical myelopathy (JOA score; Table 4). The occupying ratios of ossification (Fig. 1A) at the C3 4, C4 5, and C5 6 disc levels were calculated for the OPLL cases. The AP diameters of the spinal canal (Fig. 1B) at these levels were also measured on preoperative CT myelograms obtained in all OPLL and CSM cases involving fusions across four or more disc levels. The Mann Whitney U-test was used to analyze the incidence of C-5 paralysis depending on the occupying ratio or AP diameter (Table 5). Postoperative anterior expansion of the dura mater and anterior shift of the spinal cord were measured at the C4 5 on pre- and postoperative MR images or CT myelograms available for each patient. The distance between the posterior wall of the spinal canal and the anterior dura and posterior surface of the cord was measured with a caliper (Fig. 2A and B). The decompression width was also measured using the same images. The measurement was made by comparing the pre- and postoperative images of the same slices, and, in cases in which there was difficulty in defining the intraoperative decompression width, by using the follow-up images when the graft was completely incorporated (Fig. 2C). Intergroup statistical analysis between patients with and without C-5 palsy was performed using the Mann Whitney U-test. The relationship of the incidence of paralysis to sex and cause of disease was analyzed using the chi-square test. The statistical difference in the laterality of paralysis was analyzed by binominal distribution analysis. Spearman rank coefficient analysis was used for analysis of the number of fused segments and the AP diameters at the C3 4, C4 5, and C5 6 levels. Multiple regression analysis was performed to identify factors influencing the final MMT value. In statistical analyses significance was set at a 5% level. Operative Technique TABLE 2 Fused levels and incidence of paralysis No. of Levels Fused No. of Cases of Fused Level Cases Paralysis (%) 1 C C C C C C C C6 T C (1.7) C (3.2) C5 T C (6.3) C (6.3) C4 T C (19) C3 T C2 T1 2 0 total (3.2) The operation was performed after induction of general anesthesia, and via a left anterior approach, a discectomy or partial corpectomy was performed using microscopic guidance. The dura mater was exposed by complete or partial resection of the PLL. An anterior cervical plate was implanted in the patients in whom the fusion incorporated fewer than three disc levels (183 cases-116 with Orion plates and 67 Atlantis plates). Autologous fibula or tricortical iliac crest was used for anterior reconstruction. Philadelphia collar fixation was initiated for 2 months postoperatively. For long fusions those crossing four or more disc levels anterior corpectomy, autologous fibula graft placement, and screw fixation (the Shikata technique) were preferred (no plate was used). Instead, we ordered halo vest therapy for the first 6 weeks and Philadelphia collar fixation for the subsequent 4 weeks postoperatively. Results Radiculopathy of C-5 occurred in the 18 patients (10 with OPLL and eight with CSM; Table 3). Weakness was present in the deltoid muscles in all 18, the biceps muscle in 17, and the triceps muscles in two cases right sided, 11 left sided, and one bilaterally (Table 3). The incidence of the paralysis when examined according to sex distribution showed no significant difference 211

3 M. Ikenaga, et al. TABLE 3 Summary of demographic data obtained in 18 patients with postoperative radiculopathy* JOA Score MMT Grade Postop Case Age (yrs), Levels Impaired Onset of Months to No. Sex Lesion Fused Muscles Preop FU Laterality Preop At Onset FU Paralysis Recovery 1 49, M CSM C3 7 D, B lt days , F CSM C3 7 D, B 9 14 lt hrs , M CSM C3 7 D, B, T 6 10 lt days , M CSM C3 7 D, B lt hrs , M CSM C3 7 D, B lt days , F CSM C3 7 D, B lt hrs , F CSM C4 7 D, B rt hrs , M CSM C4 7 D, B rt day , F OPLL C3 7 D, B lt days , F OPLL C3 7 D, B 3 10 lt hrs , M OPLL C2 7 D, B, T 8 14 rt days , M OPLL C2 7 D, B bilat hrs , M OPLL C2 7 D, B 9 9 lt days , M OPLL C3 7 D, B lt hrs , M OPLL C2 6 D, B 12 lt hrs , M OPLL C3 7 D, B rt days , M OPLL C3 6 D rt days , M OPLL C4 7 D, B rt hrs 3 mean days 4.1 * B = biceps; D = deltoid; FU = follow up; T = triceps. (chi-square test, p = 0.40). Its incidence was significantly higher in patients with OPLL than in those with CSM (chi-square test, p = 0.03). Laterality of the paralysis did not show a significant difference (binominal distribution analysis, p = 0.48). There were three cases of five-level fusion, 11 cases of four-level fusion, and four cases of three-level fusion. The overall incidence of palsy was 3.2%, but the incidence increased in parallel with the number of levels fused (correlation coefficient r = 0.94) (Table 2). At onset of weakness MMT of the affected muscles showed Grade 4/5 strength in two cases, 3/5 in nine, 2/5 in four, 1/5 in two, and 0/5 in one. The Spearman rank correlation coefficients showed that the AP diameters at C3 4, C4 5, and C-5 6 were not correlated with the MMT value at onset of the paralysis. Multivariate analysis revealed that the MMT grade at onset did not relate to the age, preoperative paralysis, number of levels fused, preoperative JOA score, or severity of compression at C3 4, C4 5, and C-5 6. The onset of the muscle weakness occurred within 24 hours of surgery in nine cases, 1 day in five, 3 days in two, 7 days in one, and 8 days in one (Table 3). Clearly detailed records of the upper-extremity functional status measured immediately after the operations were not be found in two cases in which the paralysis occurred within 4 hours of surgery. Sensory disturbance was not noted in any cases, but pain around the shoulder girdle was demonstrated in three cases. In one patient, whose paralysis occurred 8 days after the surgery, dull pain around the shoulder girdle preceded the onset of muscle weakness by 1 week. Radiographic evaluation revealed that the occupying ratio at C3 4 and C4 5 was significantly higher in patients with paralysis than in those without paralysis (Table 5). The AP diameter at C3 4 and C4 5 was significantly smaller in patients with paralysis than in those without paralysis (Table 5). Postoperative MR imaging and CT myelography revealed no cases in which the decompression was insufficient. The intergroup postoperative diameter of the dura and the decompression width between patients with and TABLE 4 The JOA scoring system for cervical myelopathy upper-extremity motor function 0 impossible to eat w/ chopsticks or spoon 1 possible to eat w/ spoon but not w/ chopsticks 2 possible to eat w/ chopsticks but inadequate 3 possible to eat w/ chopsticks but awkward 4 normal lower-extremity motor function 0 impossible to walk 1 need cane or aid on flat ground 2 need cane or aid only on stairs 3 possible to walk w/o cane or aid but slowly 4 normal sensory function upper-extremity 0 apparent sensory disturbance 1 minimal sensory disturbance 2 normal lower-extremity 0 apparent sensory disturbance 1 minimal sensory disturbance 2 normal trunk 0 apparent sensory disturbance 1 minimal sensory disturbance 2 normal bladder function 0 urinary retention or incontinence 1 severe dysuria (sense of retention, dribbling, thin stream, protracted micturition) 2 slight dysuria (Pollakiuria, retarded micturition) 3 normal total score for normal

4 Radiculopathy of C-5 after anterior cervical decompression TABLE 5 Summary of occupying ratios and AP diameter stratified by patients with and without paralysis Variable w/ Paralysis w/o Paralysis occupying ratio (%) C3 4* C4 5* C AP diameter (mm) C3 4* C4 5* C *Significant at p FIG. 1. Upper: Occupying ratios of ossification were calculated for the OPLL (ossification of the posterior longitudinal ligament) cases. Lower: The AP diameters of the spinal canal were measured. Illustrative Case Case 15 Presentation and Examination. This 61-year-old man had a 5-year history of left-handed numbness, which had increased 3 months before his visit to our clinic. Physical examination showed hyperreflexia of his extremities and difficulties with fine finger motion of both hands. He also exhibited gait difficulties, which resulted in his need of a handrail when climbing stairs. Preoperative MMT demonstrated normal findings. His preoperative JOA score was 12. Plain x-radiography (Fig. 3A and B) and MR imaging (Fig. 4A and B) revealed C2 5 OPLL predominantly on the left side. Myelography and postmyelograhy CT scanning demonstrated severe cord compression secondary to OPLL (Fig. 5). The occupying ratio in the cord canal was 70 and 80% at C3 4 and C4 5, respectively. Operation. We performed anterior corpectomy and strut fibular graft placement from C-2 to C-6 (Fig. 6A and B). Severe adhesion of the ossified ligament and dura mater without paralysis were not significantly different (Mann Whitney U-test, p = 0.71 and 0.96, respectively). The extent of postoperative anterior shift of the spinal cord was significantly less in the group with paralysis compared with the group without paralysis ( and mm, respectively; Mann Whitney U-test, p 0.01) (Table 6). All patients were treated conservatively. Full recovery occurred in 13 cases. In three patients muscle strength recovered to Grade 3/5 and in two it recovered to Grade 4/5, but weakness remained in these five patients. Recovery was observed within 1 to 11 months (mean 4.1 months) after the onset of paralysis. Incomplete recovery was documented in four of 10 patients with OPLL and in one of eight patients with CSM; however, analysis revealed no statistically significant difference (chi-square test, p = 0.20). In all five cases of incomplete recovery, MMT at onset of paralysis was Grade 2/5 or less, and in all cases in which MMT showed a grade of at least 3/5 at onset, full recovery occurred over time (chi-square test, p = 0.001). Multivariate analysis revealed that the final MMT was closely related to the preoperative counterpart grade (computed t value = 4.17, p 0.01) and preoperative JOA score (computed t value = 2.75, p 0.05; Table 7). FIG. 2. The AP diameter of the dura (a), anterior shift of the spinal cord (b), and decompression width (c) were measured using the postoperative axial images. 213

5 M. Ikenaga, et al. TABLE 6 Summary of postoperative radiological measurements stratified by absence or presence of paralysis* W/ Para- W/O Para- Variable lysis (mm) lysis (mm) p Value AP dural diameter decompression width anterior shift of the spinal cord * Values are presented as the mean standard deviation. was observed intraoperatively. Careful decompression and resection were performed under the microscope. Because of the adhesion, a thin fragment of the ossified ligament was left on the dura mater to prevent leakage of cerebrospinal fluid. Postoperative Course. The numbness and fine finger motion improved postoperatively, and MMT of the upper extremities just after the operation showed Grade 5/5 strength; however, he experienced shoulder and elbow weakness 4 hours after surgery. On the following day, MMT showed Grade 1/5 strength for the deltoid and biceps muscles on his left side. Postoperative radiography and CT scanning revealed complete resection of the ossified PLL, and the decompression appeared perfect (Fig. 7 A and B). A rehabilitation program, including physiotherapy and occupational therapy, commenced on Day 1 after the operation. Active muscle exercises and passive range of motion exercises were initiated with the assistance of a physiotherapist. Passive range of motion exercises involving a pulley were also performed to prevent joint contracture. The rehabilitation programs continued for 3 months. Muscle power recovered gradually to Grade 3/5 in the deltoid and biceps muscles 3 months postoperatively. FIG. 3. Preoperative AP (left) and lateral (right) radiographs revealing C2 5 OPLL. mechanism of C-5 palsy. 19 Yonenobu and associates 25 described five of 180 patients with C-5 palsy. Chiba, et al., 2 reported 16 cases (7.7%) of paralysis in 208 patients who underwent open-door laminoplasties. Minoda, et al., 11 recently reported an incidence of paralysis of 8.9% (four of 45 patients) after midsagittal-splitting laminoplasty. There have been, however, only a few detailed reports concerning C-5 palsy after anterior cervical corpectomy and fusion. Saunders 13 reported an incidence of paralysis of 2.5% (one of 40 patients), Yonenobu and coworkers 25 reported paral- Discussion There have been several reports in which investigators have described the incidence of C-5 palsy after cervical laminoplasty. Tsuzuki and colleagues 20 have stated that C-5 nerve root palsy occurred in 13 (8.4%) of 155 patients, and later they described in detail the pathological TABLE 7 Summary of computed t values of each factor in the multivariate analysis for the final MMT grade Factors Computed t Value preop MMT grade 4.17* age 0.56 preop JOA score 2.75 time needed for recovery 0.09 AP diameter C C C no. of levels fused 1.15 duration before onset of paralysis 0.15 * Significant at p Significant at p FIG. 4. Preoperative sagittal T 1 (left) and T 2 -weighted (right) MR images demonstrating C2 5 OPLL. 214

6 Radiculopathy of C-5 after anterior cervical decompression FIG. 5. Preoperative CT myelograms of the C3 4 (upper) and C4 5 levels (lower) revealing severe cord compression secondary to the ossified ligament predominantly on the left side. ysis in eight of 205 cases, and Wada, et al., 22 reported an incidence of 4% with anterior corpectomy and fusion of 2.5 levels. In our series, the incidence of paralysis tended to increase in parallel with the number of the levels fused, but the incidence of palsy in 563 cases was 3.2% for a mean of 2.9 levels of fusion. This incidence seems to be almost identical to that reported by Wada, et al. (4.4% for a mean of 2.5 levels fused). Numerous investigators have speculated on the cause of the palsy but none has provided unequivocal evidence of its mechanism(s). In 1996, Tsuzuki, et al., 19 postulated that the tethering effect of the nerve rootlet as a mechanism caused postoperative radiculopathy after cervical laminoplasty. This theory, however, does not seem to be applicable to anterior decompression, because the spinal cord tends to move anteriorly after the procedure (Table 6), which implies possible reduction of the tensile force on the rootlet. In our series, the incidence of palsy increased with the number of levels decompressed (Table 2), underscoring the possibility that the extent of anterior dural expansion might have enhanced the incidence. Saunders 13 stated that too much anterior expansion could potentially cause an increase in root traction force. In our series, however, Spearman rank correlation testing revealed that the AP diameters at C3 4 and C4 5 had significant correlation with the number of fusion levels. Therefore, the increased incidence associated with multilevel fusion could be an indirect reflection of the severity of cord compression. Postoperative radiological studies also revealed anterior dural expansion after decompression (Table 6). The values of anterior dural expansion and the values of the decompression widths between patients with and without paralysis did not show any statistical difference. These data, therefore, do not provide sufficient evidence to support strongly Saunders idea that the restriction of the decompression width to less than 15 mm could serve as a preventive tactic for C-5 palsy. Postoperative anterior shift of the spinal cord, however, was significantly restricted in the group with paralysis (p 0.01) and this was thought to implicate intradural spinal cord adhesion as a possible cause of the C-5 palsy because anterior dural expansion with adhesion of the radix ventralis to the dura mater under restricted anterior movement of the spinal cord could generate increased tensile force on the radix ventralis. We were unable to establish ample evidence to prove this hypothesis at this point. Other authors have discussed that palsy may be caused by mechanisms other than the nerve root itself. Minoda, et FIG. 6. Postoperative AP (left) and lateral (right). Anterior C3 5 corpectomy and fusion with a C2 6 fibula graft. 215

7 M. Ikenaga, et al. FIG. 7. Postoperative CT myelograms of the C3 4 (upper) and C4 5 (lower) levels revealing a successful decompression. A thin fragment of the ossified ligament was left on the dura mater. al., 11 stated that C3 4 OPLL was associated with a higher risk of C-5 palsy. The partial disparity between the levels of the spinal cord compression and the segmental levels of neurological impairment implies that the cause of the paralysis is not restricted to the nerve root alone. We found that the cause may also potentially be related to the loss of function of the anterior radicular cells of the spinal cord s gray matter. Analysis of our data also showed that spinal cord compression was significantly more severe at the C3 4 or C4 5 levels in patients with paralysis compared with cases without paralysis. The paralysis occurred on the left side in 12 patients and on the right in seven. Although the difference was not statistically significant, the predominance of the left-sided incidence introduces the suspicion that incomplete decompression or unexpected cord or nerve root damage could be a potential cause of the paralysis. Careful assessment of postoperative CT scans revealed no evidence of insufficient decompression such as a remnant of the ossified PLL or osteophytes at the root canal. Tsuzuki, et al., 19 reported that longitudinal durotomy could be successfully used to treat radiculopathy. In our series, durotomy was not performed, but good recovery occurred in all patients. Conservative therapy seems to have its limitations, especially in cases in which muscle power is less than Grade 2/5 at onset. Durotomy may be a treatment possibility for severe cases, but we as yet have not attempted it. Although conservative treatment was chosen for our cases and the results were favorable, muscle weakness remained in several cases at follow-up examination. Multivariate analysis revealed that the MMT grade at onset of paralysis and preoperative JOA score were closely related with the final MMT grade (Table 7), and patients with a grade of 2/5 or less at onset had a high risk of poor outcome compared with those whose grade was at least 3/5. This suggests surgical treatment of C-5 palsy or prophylactic durotomy is indicated in patients with preoperative MMT-documented weakness. We were unable to identify any single predominant causative factor of C-5 root palsy. The absence of electromyography data are the limitation of the study. We are therefore now planning a new prospective study that includes pre- and postoperative electromyography, which could potentially reveal the pathogenesis of the complication more clearly. In this study, however, we did show the increased risk of C-5 palsy in cases of OPLL in which there was severe compression at C3 4 and C4 5. Our findings also made clear the possible risk of a poor result for the patients with preoperative MMT-documented weakness in the deltoid muscle. Anterior corpectomy and fusion for multilevel cervical myelopathy is somewhat challenging, but its safety and advantages when performed by a well-trained and skillful surgeon have been established. 18 Particularly in cases of cervical stenosis or OPLL with kyphotic alignment, anterior decompression is the treatment of choice. 17,26 In our study, the incidence of postoperative C-5 radiculopathy after the anterior approach was 3.2%; however, there was no deterioration in long tract signs. This incidence of palsy seems not to affect the usefulness of anterior decompression for cervical myelopathy. References 1. Bohlman HH: Cervical spondylosis with moderate to severe myelopathy. A report of seventeen cases treated by Robinson anterior cervical discectomy and fusion. Spine 2: , Chiba K, Toyama Y, Matsumoto M, Maruiwa H, Wantanabe M, Hirabayashi K: Segmental motor paralysis after expansive open-door laminoplasty. Spine 27: , Edwards CC, Heller JG, Murakami H: Corpectomy versus laminoplasty for multilevel cervical myelopathy: an independent matched-cohort analysis. Spine 27: , Hanai K, Inoue Y, Kawai K, Tago K, Itoh Y: Anterior decompression for myelopathy resulting from ossification of the posterior longitudinal ligament. J Bone Joint Surg Br 64: , Hirabayashi K, Satomi K, Ichimura S, Tanaka K, Wakano K, Toyama Y: [Complication of expansive open-door laminoplasty for ossification of the posterior longitudinal ligament in the cervical spine.] Rinsho Seikei Geka 23: , 1988 (Jpn) 6. Hirabayashi K, Watanabe K, Wakano K, Suzuki N, Satomi K, Ishii Y: Expansive open-door laminoplasty for cervical spinal stenotic myelopathy. Spine 8: ,

8 Radiculopathy of C-5 after anterior cervical decompression 7. Kirita Y, Miyazaki K, Hayashi T, Nosaka K, Shima M, Yamamura H, Tamaki S: [Clinical findings and surgical results of the ossification of the posterior longitudinal ligament in the cervical spine.] Rinsho Seikeigeka 10: , 1975 (Jpn) 8. Macdonald RL, Fehlings MG, Tator CH, Lozano A, Fleming JR, Gentili F, et al: Multilevel anterior cervical corpectomy and fibular allograft fusion for cervical myelopathy. J Neurosurg 86: , Matsuoka T, Yamaura I, Kurosa Y, Nakai O, Shindo S, Shinomiya K: Long-term results of the anterior floating method for cervical myelopathy caused by ossification of the posterior longitudinal ligament. Spine 26: , Mayr MT, Subach BR, Comey CH, Rodts GE, Haid RW: Cervical spinal stenosis: outcome after anterior corpectomy, allograft reconstruction, and instrumentation. J Neurosurg (Spine 2) 96:10 16, Minoda Y, Nakamura H, Konishi S, Nagayama R, Suzuki E, Yamao Y, et al: Palsy of the C-5 nerve root after midsagittal-splitting laminoplasty of the cervical spine. Spine 28: , Okada K, Shirasaki N, Hayashi H, Oka S, Hosoya T: Treatment of spondylotic myelopathy by enlargement of the spinal canal anteriorly, followed by arthrodesis. J Bone Joint Surg Am 73: , Saunders RL: On the pathogenesis of the radiculopathy complicating multilevel corpectomy. Neurosurgery 37: , Saunders RL, Bernini PM, Shirreffs TG, Reeves AG: Central corpectomy for cervical spondylotic myelopathy: a consecutive series with long-term follow-up evaluation. J Neurosurg 74: , Shinomiya K, Kurosa Y, Fuchioka M, Furuya K: Clinical study of dissociated motor weakness following anterior cervical decompression surgery. Spine 14: , Shinomiya K, Okawa A, Nakao K, Mochida K, Haro K, Sato T, et al: Morphology of C-5 ventral nerve rootlets as part of dissociated motor loss of deltoid muscle. Spine 19: , Suda K, Abumi K, Ito M, Shono Y, Kaneda K, Fujiya M: Local kyphosis reduces surgical outcomes of expansive open-door laminoplasty for cervical spondylotic myelopathy. Spine 28: , Tani T, Ushida T, Ishida K, Iai H, Noguchi T, Yamamoto H: Relative safety of anterior microsurgical decompression versus laminoplasty for cervical myelopathy with a massive ossified posterior longitudinal ligament. Spine 27: , Tsuzuki N, Abe R, Saiki K, Shongshi L: Extradural tethering effect as one mechanism of radiculopathy complicating posterior decompression of the cervical spinal cord. Spine 21: , Tsuzuki N, Tanaka H, Abe R, Hotta Y, Okai K: [Cervical radiculopathy occurring after the posterior decompression of the cervical spinal cord.] Rinsho Seikei Geka 26: , 1991 (Jpn) 21. Uematsu Y, Tokuhashi Y, Matsuzaki H: Radiculopathy after laminoplasty of the cervical spine. Spine 23: , Wada E, Suzuki S, Kanazawa A, Matsuoka T, Miyamoto S, Yonenobu K: Subtotal corpectomy versus laminoplasty for multilevel cervical spondylotic myelopathy: a long-term follow-up study over 10 years. Spine 26: , Whitecloud TS, LaRocca H: Fibular strut graft in reconstructive surgery of the cervical spine. Spine 1:33 43, Yonenobu K, Hosono N, Iwasaki M, Asano M, Ono K: Laminoplasty versus subtotal corpectomy: a comparative study of results in multisegmental cervical spondylotic myelopathy. Spine 17: , Yonenobu K, Hosono N, Iwasaki M, Asano M, Ono K: Neurologic complications of surgery for cervical compression myelopathy. Spine 16: , Zdeblick TA, Bohlman HH: Cervical kyphosis and myelopathy: treatment by anterior corpectomy and strut-grafting. J Bone Joint Surg Am 71: , 1989 Manuscript received March 24, Accepted in final form June 7, Address reprint requests to: Minoru Ikenaga, M.D., Ph.D., Department of Orthopedic Surgery, Kyoto City Hospital, 1-2 Higashitakada-cho, Mibu, Nakagyo-ku, Kyoto-shi, Kyoto , Japan. taketori@nyc.odn.ne.jp. 217