10:000–000, 2009 J Neurosurg Spine 10:404–413,
Surgical results of intramedullary spinal cord tumor with spinal cord monitoring to guide extent of resection Clinical article Yukihiro Matsuyama, M.D., Yoshihito Sakai, M.D., Yoshito Katayama, M.D., Shiro Imagama, M.D., Zenya Ito, M.D., Norimitsu Wakao, M.D., Koji Sato, M.D., Mitsuhiro Kamiya, M.D., Yasutsugu Yukawa, M.D., Tokumi Kanemura, M.D., Makoto Yanase, M.D., and Naoki Ishiguro, M.D. Department of Orthopaedic Surgery, Nagoya University School of Medicine, Nagoya, Japan Object. The authors investigated the outcome of intramedullary spinal cord tumor surgery, focusing on the effect of preoperative neurological status on postoperative mobility and the extent of tumor excision guided by intraoperative spinal cord monitoring prospectively. Methods. Intramedullary spinal cord tumor surgery was performed in 131 patients between 1997 and 2007. The authors compared the pre- and postoperative neurological status and examined the type of surgery in 106 of these patients. A modified McCormick Scale (Grades I–V) was used to assess ambulatory ability (I = normal ambulation; II = mild motor sensory deficit, independent without external aid; III = independent with external aid; IV = care required; and V = wheelchair required). The type of surgery was classified into 4 levels: total resection, subtotal resection, partial resection, and biopsy. Results. The 106 patients consisted of 47 females and 59 males, whose average age was 42.5 years (range 6–75 years). The mean follow-up period was 7.3 years (range 2.5 months–21 years). The tumor types included astrocytoma (12 cases), ependymoma (46 cases), hemangioblastoma (16 cases), cavernous hemangioma (17 cases), and others (15 cases overall: gangliocytoma, 1; germ cell tumor, 1; lymphoma, 3; neurinoma, 1; meningioma, 1; oligodendroglioma, 1; sarcoidosis, 2; glioma, 1; and unknown, 4). Initial total excision, subtotal resection, partial resection, biopsy, and duraplasty were performed in 59, 12, 22, 12, and 1 patients, respectively. According to the preoperative McCormick Scale, ambulatory status was classified as Grades I, II, III, IV, and V in 41(38%), 30 (28%), 14 (13%), 19 (19%), and 2 (2%) patients, respectively. Thirty-three (31%) of 106 patients suffered postoperative neurological deterioration. The number of patients who did not lose ambulatory ability or who achieved an ambulatory status of Grade I or II postoperatively was 33 (80%), 21 (70%), 10 (71%), 8 (42%), and 1 (50%) in patients with preoperative Grades I, II, III, IV, and V, respectively. Total excision was performed in 31 (79%) of 39 patients with preoperative Grade I, 12 (40%) of 30 patients with Grade II, 7 (50%) of 14 patients with Grade III, and 9 of 21 patients (38%) with Grade IV or V, indicating that the rate of total excision was significantly higher in patients with Grade I status. Conclusions. The postoperative ambulatory ability was excellent in patients with a good preoperative neurological status. Total excision in patients with Grade I or II ambulation was associated with a good prognosis for postoperative mobility. However, the rate of postoperative deterioration was 31.5%, which is relatively high, and patients should be fully informed of this concern prior to intramedullary spinal cord tumor surgery. (DOI: 10.3171/2009.2.SPINE08698)
Key Words • intramedullary tumor • spinal cord tumor • spinal cord monitoring
I
spinal cord tumors are very rare, accounting for 2–4% of CNS tumors.1,3,9,10,13 Unlike brain tumors, most cases, with a few exceptions, are pathologically benign, but surgical treatment is difficult. Since Epstein et al.6,7 reported the outcomes of active resection of IMSCT in 1980, modern microscopic techniques have developed and many facilities have reported ntramedullary
Abbreviations used in this paper: CMAP = compound muscle action potential; IMSCT = intramedullary spinal cord tumor; MEP = motor evoked potential; SSEP = somatosensory evoked potential.
404
surgical outcomes. Neurological manifestations were aggravated postoperatively in > 30% of the cases in many reports,4,5,8,11,12,19,26,27 showing that the surgery is still challenging. Moreover, since IMSCTs are very rare, only a few cases are encountered in each facility. Accordingly, only a few studies have investigated the extent of tumor resection and optimum timing of surgery based on the nervous condition. In this study, we investigated the outcomes of IMSCT surgery, particularly the influences of the preoperative gait status on the postoperative gait and the amount of tumor resection, to determine the optimum timing of surgery. J. Neurosurg.: Spine / Volume 10 / May 2009
Surgical results of intramedullary spinal cord tumor Methods
Of 131 patients with IMSCT surgically treated at our hospital and related hospitals between 1997 and 2007, pre- and postoperative gait status and the extent of tumor resection were investigated in 106 patients (59 males and 47 females) whose mean age was 42.5 years (range 6–75 years). The mean follow-up duration was 7.3 years (range 2.5 months–21 years). The pathological diagnosis was ependymoma in 46, astrocytoma in 12, hemangioblastoma in 16, cavernous hemangioma in 17, and other in 15 (gangliocytoma, 1; germ cell tumor, 1; lymphoma, 3; neurinoma, 1; meningioma, 1; oligodendroglioma, 1; sarcoidosis, 2; glioma, 1; and unknown, 4). The initial surgery was total resection in 59, subtotal resection in 12, partial resection in 22, biopsy in 12, and duraplasty alone in 1. Ambulatory ability was evaluated using the modified McCormick Scale19,20 (grade range I–V: I = normal gait, II = mild gait disturbance not requiring support, III = gait with support, IV = assistance required, and V = wheelchair needed). The evaluation was performed at 3 weeks and at 1 year after surgery. The amount of tumor resection was evaluated by categorizing into the following 4 steps: total resection, subtotal resection, partial resection, and biopsy. In all operative cases, total resection was attempted. We used the standard definition of total resection: removal of 100% of the tumor as evidenced by both a microscopically documented clean surgical field at the end of the procedure and intraoperative ultrasonography-documented clean surgical bed. When a small tumor fragment was deliberately left in place, the procedure was considered to be a subtotal resection, based on documented intraoperative ultrasonography of 80–99%. We performed subtotal resection in this series when intraoperative evoked potential monitoring changes heralded impending neurological paralysis. In the same manner, we defined 50–80% resection as partial resection and < 50% resection was defined as a biopsy. The extent of resection was quantified using ultrasonography and direct visualization with microscope. Also, we confirmed the extent of resection by using MR imaging 1 month postoperatively. Despite their individual tumor grades, all patients underwent the same intraoperative evoked potential monitoring. Surgical Procedure and Spinal Cord Monitoring
All surgeries were performed using a microscope, and the IMSCT was approached via the midline of the spinal cord. In some cases of deviated hemangioblastoma and hemangioma, a peritumoral approach was used instead of the midline approach. Just before the midline approach to the pia mater, 200 ml of mannitol was administered intravenously for 10 minutes to decrease spinal cord edema. Whether the spinal tumor resection procedure should be performed was determined based on spinal cord monitoring. We used CMAP monitoring2,14–17,21–25 as the intraoperative spinal cord modality, and when the CMAP waveform worsened during the procedure (multiple-phase waveform or loss of wave) the manipulation of the spinal cord tumor was suspended and resumed after recovery of
J. Neurosurg.: Spine / Volume 10 / May 2009
the waveform. When the waveform did not recover, total tumorectomy was discontinued. When the waveform recovered and then worsened again after reinitiation of the surgical procedure, the procedure was discontinued even when total resection could not be completed. Eighty to 90% of the tumor had been excised when the waveform was aggravated in most cases, and this situation occurred during the final dissection between the ventral region of the tumor and spinal cord. In patients with preoperative McCormick grades of III–V, the CMAPs were sometimes an atypical multiple-phase wave or were undetectable. In such cases, we defined this atypical multiple-phase wave or normal detectable wave in 16 other muscles monitored as a control wave; if this wave was changed or lost, this was taken as a warning. After tumor resection, intraoperative ultrasonography was used to confirm whether any residual tumor was present, and the dura mater was sutured using Gore-Tex to prevent adhesion between the manipulated spinal cord and dura. We used frozen sections for pathological examination, especially for astrocytoma. Depending on the pathology (that is, anaplastic astrocytoma), the dissection was terminated even if the waveforms were stable. Monitoring Conditions
Transcranial stimulation was performed with a Multiphase Stimulator D185 (Digitimer). The stimulus conditions were 4–5 train stimuli, stimulus interval 2 msec, stimulus 300–600 V, filter 50–1000 Hz, recording time 100 msec, and total of < 20 stimuli. The stimulator was 2 cm anterior and 3 cm lateral to Cz (International 10–20 system) over the cerebral cortex motor area. Using Neuropack MEB-2200 (version 04.02), which is expandable to 16 channels, MEPs were recorded from the peripheral limbs via needle and disc (diameter 6 mm) electrodes, and from the anus via plug-type electrodes. The evoked muscles, depending on the site of surgery, were the deltoid, biceps, triceps, hypothenar, quadriceps femoris, hamstrings, tibialis anterior, gastrocnemius, and sphincters.
Anesthesia Protocol
Because benzodiazepine, as a preanesthetic medication, suppresses latency and amplitude, it was either not used or minimized. The drugs administered were propofol (3–4 µg/ml), fentanyl (2 µg/kg), and vecuronium (0.12– 0.16 mg/kg). Anesthesia was maintained using propofol (3 µg/ml), fentanyl (1–2.5 µg/kg/h), and vecuronium (0.01– 0.04 mg/kg/h). Concomitant hypotensive anesthesia was given as appropriate with continuous administration of PGE1 and a short-acting β1 blocker (landiolol). Free-Running Electromyography
Real-time monitoring with constant free-running electromyography was performed intraoperatively, especially during critical phases. Meanwhile, during rest only the baseline or a small, irregular spontaneous discharge (amplitude 100–200 µV, duration ~ 3 msec), called an 405
Y. Matsuyama et al. Table 1a: Summary of immediate and 1-year McCormick grades measured in overall population of 106 patients with IMSCTs Postop McCormick Grade Grade I II III IV V total
1-Yr McCormick Grade
Worsening Gait
Smooth Gait
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
(%)
at 1 Yr
(%)
at 3 Wks
(%)
22 9 3 2 0 36
11 12 7 6 1 37
4 4 3 2 1 14
1 3 0 4 0 8
3 2 1 5 0 11
29 11 4 2 0 46
7 13 7 6 2 35
2 5 2 5 0 14
3 1 1 4 0 9
0 0 0 2 0 2
41 30 14 19 2 106
19 9 1 5 0 34
(46) (30) (7) (26) (0) (32)
12 6 1 2 0 21
(29) (20) (7) (10) (0) (20)
33 21 10 8 1 73
(80) (70) (71) (42) (50) (69)
endplate spike, was observed; once damage was inflicted on the spine or nerve root, a high-amplitude, long-duration motor unit potential was observed. If this waveform was observed, the surgical maneuvers were temporarily ceased and monitoring was performed with CMAP under the aforementioned conditions. For statistical analysis, repeated analysis of variance was used, and p < 0.05 was regarded as significant.18
Results All Tumors
There were 106 tumors overall (Tables 1a and b). The preoperative McCormick Scale grade was I in 41 (38%), II in 30 (28%), III in 14 (13%), IV in 19 (19%), and V in 2 (2%) patients. Neurological status was aggravated after surgery in 34 (32%) of the 106 patients. In patients who could not walk after surgery and whose McCormick Scale grade was IV or V, the preoperative score was I in 4 patients (10%), II in 5 (17%), III in 1 (7%), and IV and V in 9 (45%). Gait disturbance was almost absent after surgery—Grade I or II in 33 (80%) of the 41 cases with Grade I status before surgery, 21 (70%) of the 30 with preoperative Grade II, 10 (71%) of the 14 with preoperative Grade III, 8 (42%) of the 19 with preoperative Grade IV, and 1 (50%) of the 2 with preoperative Grade V. On
at 1 Yr 36 24 11 8 2 81
(%) (88) (80) (78) (42) (100) (76)
statistical analysis, postoperative gait was significantly better when the preoperative McCormick scale was I–III (p < 0.001). At 1 year after surgery, 36 (88%) of the 41 preoperative Grade I cases, 24 (80%) of the 30 preoperative Grade II cases, 11 (78%) of the 14 preoperative Grade III cases, and 8 (42%) of the 19 preoperative Grade IV cases had almost no gait disturbance (Grade I or II), indicating that, when the preoperative grade was I–III, the gait was favorable immediately to 1 year after surgery and that the gait was improved at 1 year, compared with that immediately after surgery (Table 1a). Regarding the relationship between the amount of tumor resection and preoperative neurological manifestations, total resection was performed in 31 (76%) of the 41 patients with preoperative Grade I status, 12 (40%) of the 30 patients with preoperative Grade II status, 7 (50%) of the 14 patients with preoperative Grade III status, and 9 (38%) of the 21 patients with preoperative Grade IV status, demonstrating that the total resection rate was significantly higher in patients with good preoperative gait (Grade I) (p < 0.001) (Table 1b). The outcomes were investigated in individual pathologies, as discussed below. Ependymoma
There were 27 male and 19 female patients with ependymomas (mean age 44.4 years, range 6–75 years)
Table 1b: Correlation of McCormick grade and extent of tumor resection in overall population of 106 patients with IMSCTs Extent of Resection McCormick Grade
Subtotal
Partial
Biopsy
Total No. of Cases (%)
31
3
4
1
39 (79)
II
12
5
10
3
30 (40)
III
7
1
3
3
14 (50)
IV
7
3
5
5
20 (35)
V
2
0
0
1
3 (67)
59
12
22
13
106 (56)
total
406
Total
I
J. Neurosurg.: Spine / Volume 10 / May 2009
Surgical results of intramedullary spinal cord tumor Table 2a: Summary of immediate and 1-year McCormick grades in 46 patients with ependymomas Postop McCormick Grade Grade I II III IV V total
1-Yr McCormick Grade
Worsening Gait
Smooth Gait
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
(%)
at 1 Yr
(%)
at 3 Wks
12 3 1 1 0
6 2 4 4 0
1 2 1 1 0
1 1 0 2 0
2 0 0 2 0
16 4 1 1 0
2 2 5 4 0
2 2 0 1 0
2 0 0 3 0
0 0 0 1 0
22 8 6 10 0
10 3 0 2 0
(45) (36) (0) (20) (0)
6 2 0 1 0
(27) (25) (0) (10) (0)
18 5 5 5 0
(82) (63) (83) (50) (0)
18 6 6 5 0
(82) (75) (100) (50) (0)
17
16
5
4
4
22
13
5
5
1
46
15
(33)
9
(19)
33
(72)
35
(76)
(Tables 2a and b). The level of the tumor location was the cervical region in 21, the cervicothoracic in 8, the thoracic region in 8, and the conus medullaris region in 9. The mean duration of follow-up was 7.2 years (range 1–19 years). The extent of tumor resection was total in 30, subtotal in 8, partial in 8, and biopsy in 0, and the total resection rate was 65%. Including subtotal resection, the tumor was resectable in 83%. Neurological status was aggravated in 15 (33%) of the 46 patients, and 33 (72%) could easily walk after surgery (McCormick Scale Grade I or II). Of 22 patients with preoperative Grade I status, 18 could walk (82%), showing a significantly high rate (p < 0.001). At 1 year after surgery, 18 (82%) of 22 preoperative Grade I cases, 6 (75%) of 8 preoperative Grade II cases, 6 of 6 preoperative Grade III cases, and 5 (50%) of 10 preoperative Grade IV cases were Grade I or II with almost no gait disturbance, indicating that, when the preoperative McCormick Scale grade was I–III, gait was significantly better immediately and 1 year after surgery and that gait was improved at 1 year, compared with that immediately after surgery (Table 2a). Beyond the 1st-year evaluation, 3 of 8 patients with partial tumor resections suffered neurological deterioration and underwent reoperation. The first case involved a thoracic ependymoma, and the patient’s neurological status was Grade II preoperatively, aggravated to Grade IV immediately after surgery, and recovered to Grade III
(%)
at 1 Yr
(%)
after 1 year. Eight years after the surgery, the patient’s neurological status had deteriorated to Grade IV due to enlargement of tumor. This patient underwent a subtotal resection and status improved to Grade III. The second case also involved a thoracic ependymoma, and the patient’s neurological status was Grade II preoperatively, aggravated to Grade III immediately after the surgery, and recovered to Grade II after 1 year. Ten years after surgery, the patient’s neurological status had deteriorated to Grade IV due to enlargement of tumor. This patient underwent a subtotal resection and status improved to Grade II. The third case was a conus region ependymoma and its neurological status was Grade IV preoperatively, aggravated to Grade V immediately after the surgery and recovered to Grade IV after 1 year. Six years after the surgery, neurological status had deteriorated to Grade V due to enlargement of tumor. This patient underwent surgery with subtotal resection and improved to Grade VI. Regarding the relationship between the amount of tumor resection and preoperative McCormick Scale grade, total resection was possible in 18 (82%) of the 22 preoperative Grade I cases, indicating that the total resection rate was significantly higher when surgery was performed in Grade I cases than in cases involving other grades (p < 0.001) (Table 2b). Astrocytoma
There were 7 male and 5 female patients (mean age
Table 2b: Correlation of McCormick grade and extent of tumor resection in 46 patients with ependymomas Extent of Resection McCormick Grade
Total
Subtotal
Partial
Biopsy
Total No. of Cases (%)
I
18
3
1
0
22 (82)
II
4
1
3
0
8 (50)
III
4
1
1
0
6 (67)
IV
4
3
3
0
10 (40)
V
0
0
0
0
0 (0)
30
8
8
0
46 (65)
total
J. Neurosurg.: Spine / Volume 10 / May 2009
407
Y. Matsuyama et al. Table 3a: Summary of immediate and 1-year McCormick grades in 12 patients with astrocytomas Postop McCormick Grade
1-Yr McCormick Grade
Worsening Gait
Grade
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
I II III IV V total
2 1 0 0 0 3
0 3 0 0 0 3
0 0 1 1 0 2
0 0 0 0 0 0
1 0 0 3 0 4
2 1 0 0 0 3
0 3 0 0 0 3
0 0 1 2 0 3
1 0 0 1 0 2
0 0 0 1 0 1
3 4 1 4 0 12
1 0 0 3 0 4
39.4 years, range 13–64 years) (Tables 3a and b). The level of tumor location was the cervical region in 2 patients, the cervicothoracic region in 2, the thoracic region in 3, and the conus medullaris region in 5. The mean duration of follow-up was 12.4 years (range 4.2–16.5 years). The extent of tumor resection was total in 1, subtotal in 1, partial in 3, and biopsy in 7, and the total resection rate was 8%. Neurological symptoms were aggravated in 4 (33%) of the 12 cases, and 6 (50%) could easily walk after surgery (McCormick Scale Grade I or II). In 7 preoperative Grade I and II cases, 6 patients (86%) could walk, indicating a significantly higher rate (p < 0.001). At 1 year after surgery, 6 patients had almost no gait disturbance (Grade I and II), showing no change, but in 2 of 3 patients with gait disturbances that had become aggravated to Grade V immediately after surgery, grades improved to III and IV. Thus, gait was aggravated 1 year after surgery in 2 (17%) of the 12 cases (Table 3a). Beyond the 1st-year evaluation, 1 of 3 patients with partial resections had neurologically deteriorated and underwent reoperation. This patient harbored a thoracic pilocytic astrocytoma. Neurological status was Grade III preoperatively, stable immediately after surgery, and stable still (Grade III) after 1 year. Nine years after the surgery, neurological status had deteriorated to Grade IV due to enlargement of tumor. This patient underwent subtotal resection and status improved to Grade III. Regarding the relationship between the extent of tumor resection and preoperative McCormick Scale grade,
(%)
at 1 Yr
(33) (0) (0) (75) (0) (33)
1 0 0 1 0 2
Smooth Gait (%)
at 3 Wks
(%)
at 1 Yr
(%)
(33) (0) (0) (25) (0) (17)
2 4 0 0 0 6
(67) (100) (0) (0) (0) (50)
2 4 0 0 0 6
(67) (100) (0) (0) (0) (50)
partial resection or biopsy was performed in 10 of the 12 cases, and total resection was performed in only 1 case. Seven of the lesions were of a fibrillar-type, highly anaplastic tumor. In many cases intraoperative pathological diagnosis and dissection were difficult due to the unclear boundary between the tumor and normal spinal cord, and only biopsy was performed. Six of 7 patients with anaplastic astrocytoma died within 1.5 years after the onset (Table 3b). Hemangioma
There were 7 male and 10 female patients with a mean age of 41.6 years (8–66 years) (Tables 4a and b). The spinal level of tumor location was cervical in 6, cervicothoracic in 2, thoracic in 8, and conus medullaris region in 1. The mean duration of follow-up was 5.5 years (range 2–10 years). The extent of tumor resection was total in 12 patients, subtotal in 2, partial in 3, and biopsy in none, and the total resection rate was 70% (12 of 17 cases). Including subtotal resection, resection was possible in 83% of these cases. Neurological symptoms were aggravated in 5 (29%) of the 17 cases, and 15 (89%) of the 17 patients could easily walk after surgery (McCormick Scale Grade I or II). In 7 (100%) of 7 preoperative Grade I cases, 6 (88%) of 7 preoperative Grade II cases, and 1 of 1 preoperative Grade III case, the patient was able to walk with ease after surgery, showing a significantly high rate (p < 0.001). At 1 year after surgery, 2 of 4 patients in whom
Table 3b: Correlation of McCormick grade and extent of tumor resection in 12 patients with astrocytomas Extent of Resection McCormick Grade
408
Total
Subtotal
Partial
Biopsy
Total No. of Cases (%)
I
0
0
1
2
3 (0)
II
1
1
0
2
4 (25)
III
0
0
1
0
1 (0)
IV
0
0
1
3
4 (0)
V
0
0
0
0
0 (0)
total
1
1
3
7
12 (8)
J. Neurosurg.: Spine / Volume 10 / May 2009
Surgical results of intramedullary spinal cord tumor Table 4a: Summary of immediate and 1-year McCormick grades in 17 patients with hemangiomas Postop McCormick Grade
1-Yr McCormick Grade
Worsening Gait
Smooth Gait
Grade
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
(%)
at 1 Yr
(%)
at 3 Wks
(%)
at 1 Yr
(%)
I II III IV V total
3
4
0
0
0
5
2
0
0
0
7
4
(57)
2
(29)
7
(100)
7
(100)
2 1 0 0 6
4 0 0 1 9
0 0 0 1 1
0 0 0 0 0
1 0 0 0 1
3 1 0 0 9
3 0 0 2 7
1 0 0 0 1
0 0 0 0 0
0 0 0 0 0
7 1 0 2 17
1 0 0 0 5
(14) (0) (0) (0) (29)
1 0 0 0 3
(14 (0) (0) (0) (18)
6 1 0 1 15
(86) (100) (0) (50) (88)
6 1 0 2 16
(86) (100) (0) (100) (94)
symptoms had deteriorated to Grade II immediately after surgery recovered to Grade I, and the deterioration rate was reduced to 18% (Table 4a). Beyond the 1st year evaluation, 1 of 3 patients with a partial resection had deteriorated neurologically and underwent reoperation. In this case the patient harbored a thoracic cavernous hemangioma, and neurological status was Grade II preoperatively, aggravated to Grade V immediately after surgery, and recovered to Grade III after 1 year. Seven years after the surgery, neurological status had deteriorated to Grade IV due to intratumoral bleeding. This patient underwent total resection and neurological status improved to Grade III. Regarding the relationship between the extent of tumor resection and preoperative McCormick grade, total resection was possible in 9 (90%) of the 10 preoperative Grade I cases, and the total resection rate was significantly higher at Grade I than at Grade II (p < 0.001) (Table 4b). Hemangioblastoma
There were 9 male and 7 female patients with a mean age of 41 years (21–68 years) (Tables 5a and b). The spinal level of tumor location was cervical in 7, cervicothoracic in 2, thoracic in 5, and the conus medullaris region in 2. The mean duration of follow-up was 5.5 years (range 2–20 years). The extent of tumor resection was total in 13, subto-
tal in 0, partial in 3, and biopsy in 0, and the total resection rate was 81% (13 of 16 cases). Neurological findings were aggravated in 7 (47%) of the 16 cases, and 8 (50%) of the 16 patients could easily walk after surgery (Grade I or II). In 4 (57%) of 7 preoperative Grade I cases, 3 (50%) of 6 preoperative Grade II cases, and 1 of 1 preoperative Grade III case, the patient could walk easily after surgery, indicating no significant difference between the pre- and postoperative gait conditions. At 1 year after surgery, 7 of the 7 patients with preoperative Grade I status and 5 (83%) of the 8 with preoperative Grade II status had almost no gait disturbance (Grade I or II). One and 2 of 3 cases that deteriorated to Grade III immediately after surgery were improved to Grades I and II, respectively, and 82% of the patients could walk easily at 1 year (Table 5a). Beyond the 1st-year evaluation, all 3 patients with partial resections had neurological deterioration and underwent reoperation. In the first case the patient harbored a cervical hemangioblastoma, and neurological status was Grade II preoperatively, aggravated to Grade IV immediately after surgery, and recovered to Grade III after 1 year. Two years after surgery, neurological status had deteriorated to Grade IV due to enlargement of a syrinx in the spinal cord. This patient underwent total resection and status improved to Grade III. In the second case the patient harbored a thoracic hemangioblastoma, and neurological status was Grade II preoperatively, deteriorated
Table 4b: Correlation of McCormick grade and extent of tumor resection in 17 patients with hemangiomas Extent of Resection McCormick Grade
Total
Subtotal
Partial
Biopsy
Total No. of Cases (%)
I
9
0
1
0
10 (90)
II
1
2
2
0
5 (20)
III
1
0
0
0
1 (100)
IV
0
0
0
0
0 (0)
1
0
0
0
1 (0)
12
2
3
0
17 (71)
V total
J. Neurosurg.: Spine / Volume 10 / May 2009
409
Y. Matsuyama et al. Table 5a: Summary of immediate and 1-year McCormick grades in 16 patients with hemangioblastomas Postop McCormick Grade
1-Yr McCormick Grade
Worsening Gait
Grade
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
I II III IV V total
4
0
3
0
0
5
2
0
0
0
7
2 0 1 0 7
1 0 0 0 1
2 1 0 0 6
1 0 0 0 1
0 1 0 0 1
2 0 1 0 8
3 0 0 0 5
1 1 0 0 2
0 1 0 0 1
0 0 0 0 0
6 2 1 0 16
to Grade III immediately after surgery, and recovered to Grade II after 1 year. One and a half years after the surgery, neurological status deteriorated to Grade III due to enlargement of a syrinx in the spinal cord. This patient underwent total resection, and status improved to Grade II. In the third case the patient harbored a cervical hemangioblastoma, and neurological status was Grade III preoperatively, deteriorated to Grade V immediately after surgery, and recovered to Grade IV after 1 year. Two years after the surgery, neurological status had deteriorated to Grade V due to enlargement of a syrinx in the spinal cord. This patient underwent total resection and status improved to Grade IV. Regarding the relationship between the extent of tumor resection and preoperative McCormick grade, total resection was possible in 7 of the 7 patients with a preoperative grade of I, 4 (70%) of the 6 with a preoperative grade of II, and 1 (50%) of the 2 with a preoperative grade of III, underscoring that the total resection rate was significantly higher when surgery was performed at Grade I (p < 0.001) (Table 5b). Miscellaneous Tumors
The pathological type was gangliocytoma in 1, germ cell tumor in 1, lymphoma in 3, neurinoma in 1, meningioma in 1, oligodendroglioma in 1, sarcoidosis in 2, glioma in 1, and unknown in 4 (Tables 6a and b). There were 7 male and 8 female patients with a mean
Smooth Gait
(%)
at 1 Yr
(%)
at 3 Wks
(%)
at 1 Yr
(%)
3
(43)
2
(28)
4
(57)
7
(100)
3 1 0 0 7
(50) (50) (0) (0) (47)
1 1 0 0 4
(17) (50) (0) (0) (25)
3 0 1 0 8
(50) (0) (100) (0) (50)
5 0 1 0 13
(83) (0) (100) (0) (82)
age of 36 years (range 5–68 years). The spinal level of tumor location was cervical in 6, cervicothoracic in 2, thoracic in 5, and conus medullaris in 2. The mean duration of follow-up was 6.2 years (range 2–9 years). The extent of tumor resection was total in 3, subtotal in 1, partial in 5, and biopsy in 6, and the total resection rate was 20% (3 of 15). Neurological symptoms were aggravated in 3 (20%) of the 15 cases, and 11 (73%) of the 15 patients could easily walk after surgery (Grade I or II). In 2 of 2 preoperative Grade I cases, 3 (60%) of 5 preoperative Grade II cases, 4 of 4 preoperative Grade III cases, and 2 (50%) of 4 preoperative Grade IV cases, the patient could walk easily after surgery, indicating no significant relationship between the pre- and postoperative gait. No changes were noted in these neurological symptoms after 1 year (Table 6a). Regarding the extent of tumor resection and the preoperative McCormick grade, total resection was performed in 3 patients, and subtotal in 1, but the procedure was ended at partial resection or biopsy in most cases, demonstrating no correlation (Table 6b). We excluded 12 astrocytomas and 15 miscellaneous tumors from 106 cases and evaluated the surgical results of 79 cases including 46 ependymomas, 17 hemangiomas, and 16 hemangioblastomas. In 27 of (34%) 79 cases status deteriorated immediately after surgery, although this decreased to 20% because 11 of the 27 aggravated cases recovered by 1 year. In terms of neurological aggravation
Table 5b: Correlation of McCormick grade and extent of tumor resection in 16 patients with hemangioblastomas Extent of Resection McCormick Grade
Total
Subtotal
Partial
Biopsy
I
7
0
0
0
7 (100)
II
4
0
2
0
6 (67)
III
1
0
1
0
2 (50)
IV
1
0
0
0
1 (100)
V
0
0
0
0
0 (0)
13
0
3
0
16 (81)
total
410
Total No. of Cases (%)
J. Neurosurg.: Spine / Volume 10 / May 2009
Surgical results of intramedullary spinal cord tumor Table 6a: Summary of immediate and 1-year McCormick grades in 15 patients with miscellaneous tumor types Postop McCormick Grade
1-Yr McCormick Grade
Worsening Gait
Smooth Gait
Grade
I
II
III
IV
V
I
II
III
IV
V
Total
at 3 Wks
(%)
at 1 Yr
(%)
at 3 Wks
(%)
at 1 Yr
(%)
I II III IV V total
1
1
0
0
0
1
1
0
0
0
2
1
(50)
1
(50)
2
(100)
2
(100)
1 1 0 0 3
2 3 2 0 8
0 0 0 0 0
1 0 2 0 3
1 0 0 0 1
1 2 0 0 4
2 2 2 0 7
1 0 2 0 3
1 0 0 0 1
0 0 0 0 0
5 4 4 0 15
2 0 0 0 3
(40) (0) (0) (0) (0)
2 0 0 0 3
(40) 0 0 (0) (20)
3 4 2 0 11
(60) (100) (50) (0) (73)
3 4 2 0 11
(60) (100) (50) (0) (73)
rate, there was no statistical difference between the overall 106 consecutive cases and selected 79 cases. In terms of gait disturbance, 56 (71%) of 79 patients were able to walk easily 1 month after surgery and 64 (81%) of 79 patients could do so 1 year after the surgery. As for ambulatory grade, in 29 (81%) of the 36 Grade I cases, 14 (67%) of the 21 preoperative Grade II cases, 6 (67%) of the 9 preoperative Grade III cases, 6 (55%) of the 11 preoperative Grade IV cases, and 1 (50%) of the 2 preoperative Grade V cases, the patients were able to walk easily after 1 month. One year after the surgery, in 32 (89%) of the 36 Grade I cases, 17 (81%) of the 21 Grade II cases, 7 (78%) of the 9 Grade III cases, 6 (55%) of the 11 Grade IV cases, and 1 (50%) of the 2 Grade V, the patients were able to ambulate with ease. On statistical analysis, the postoperative gait status was significantly better when the preoperative McCormick grade was I–III (p < 0.001). When the preoperative grade was I–III, the gait status was favorable immediately to 1 year after surgery, and the gait was improved at 1 year, compared with that immediately after surgery. In comparing the overall group of 106 patients and this selected group of 79 patients, we could find no difference in terms of neurological aggravation rate or ambulatory rate pertaining to smooth gait. Regarding the relationship between the extent of tumor resection and preoperative neurological manifestation, total resection was performed in 34 (87%) of the 39 patients with preoperative Grade I status, 9 (47%) of the
19 with preoperative Grade II status, 6 (67%) of the 9 with preoperative Grade III status, and 5 (45%) of the 11 with preoperative Grade IV status, demonstrating that the total resection rate was significantly higher in cases involving a good preoperative gait (Grade I) (p < 0.001). Compared with 106 cases overall, the total excision rate was significantly better in cases involving preoperative Grades I, II, and III.
Discussion
The advent of modern microsurgical techniques has led to a significant change in the approach to IMSCTs. The pioneering work of Epstein and colleagues6,7,28 and the legacy of detailed neurosurgical studies over the last 24 years have dramatically improved the outcome in patients harboring these tumors. In cases of IMSCTs one should not wait for the onset of clinical deterioration but rather institute treatment as soon as possible. The earlier the diagnosis and the more radical the resection of an IMSCT, the greater the likelihood of preserving the patient’s neurological function.28 Although the earliest possible surgical treatment is reportedly better, there have been no reports to detail the preoperative gait grade in a patient treated with surgery, his specific postoperative ambulatory ability, or the extent of tumor resection in such cases. We followed 106 surgically treated patients with IMSCTs, investigated the outcomes, particularly the post-
Table 6b: Correlation of McCormick grade and extent of tumor resection in 15 patients with miscellaneous tumor type Extent of Resection McCormick Grade
Total
Subtotal
Partial
Biopsy
Total No. of Cases (%)
I
0
0
1
0
1 (0)
II
0
1
3
1
5 (0)
III
1
0
0
3
4 (25)
IV
2
0
1
2
5 (40)
V
0
0
0
0
0 (0)
total
3
1
5
6
15 (20)
J. Neurosurg.: Spine / Volume 10 / May 2009
411
Y. Matsuyama et al. operative gait in relation to the various preoperative gait states, and identified the extent of tumor excision guided by intraoperative spinal cord monitoring. In 1999 Kane et al.13 reported that the gait status was aggravated and unchanged in 6 (12%) and 45 (82%) of 54 patients with intramedullary tumors. In 1994 Cristante and Herrmann5 reported that the outcome was aggravated, unchanged, and improved in 31, 55, and 17% of 69 intramedullary tumor cases, respectively. Constantini et al.3 reported the outcomes of aggravated, unchanged, and improved in 23.8, 60.4, and 15.8% of 164 cases of pediatric intramedullary tumor, respectively, and Sandalcioglu et al.26 reported that the outcome was aggravated in 27 (34.6%) and unchanged in 51 (65%) of 78 cases of intramedullary tumor. Despite advances in the surgical procedure, gait status became aggravated following surgery for intramedullary tumor in 20–30%. The tumor location, size, pathological type, age, preoperative neurological manifestation, and gait status were reported to affect the postoperative gait, but no report has closely described preoperative and postoperative gait 1 month and 1 year after surgery. Moreover, there has been no report correlating the extent of tumor resection with the preoperative gait. We have been performing tumor resection with intraoperative spinal cord monitoring. We made it a rule to suspend surgery when the waveform becomes aggravated, and we continue when the waveform has improved, but abandon the resection when the waveform becomes multiphasic again or is lost. As a result, the gait was aggravated and improved immediately after surgery in 34 (32%) and 31 (29%) cases, respectively, and in 21 (22%) and 37 (35%) cases after 1 year, respectively, which may be better than rates in other reports. The surgical outcomes vary depending on the type of intraoperative spinal cord monitoring, and the waveform pattern is considered a critical point, based on which continuation or discontinuation of surgery is decided. In 2006 Sala et al.23,24 monitored SSEPs and MEPs, setting the critical point of SSEPs to a 50% reduction of the amplitude or setting an alarm point of MEPs to loss of the waveform; surgery was continued when the D-wave did not decrease over 50% and abandoned when the D-wave decreased by more than 50% with loss of MEP. As a result, the outcome worsened early after surgery in 19 (38%) of 50 cases in the SSEP monitoring group, and in 15 (30%) of 50 cases in the MEP monitoring group, underscoring that better surgical outcomes were achieved using MEPs and 50% reduction of D-waves as indices, rather than SSEPs. In 2007 Kothbauer and colleagues14–16 also reported setting the critical point to loss of MEP waveform and a 50% reduction of the D-wave. Monitoring the motor pathway, CMAP, is used as the optimum intraoperative spinal cord monitoring system in many cases, but its critical point has not been established.2,14–17,21–25 We have been performing surgery with CMAP monitoring in 16 muscles, considering the change of the waveform to a multiphasic pattern or loss in many muscles as an alarm point, and we have achieved favorable surgical outcomes in comparison with those reported by others. Regarding the amount of tumor resection, total excision was possible in 59 (56%) of the 106 cases and subto412
tal resection in 12 (11%). When surgery was performed in patients with Grade I gait, total resection was achieved in 31 (79%) of 41 cases, which was satisfactory. Total resection was not achievable at a relatively high rate in cases of anaplastic-type astrocytoma, in which the boundary between the tumor and spinal cord was unclear in 7 (58%) of the 12 cases. Total resection was not possible in 12 of the 15 patients with miscellaneous tumors, which may have also decreased the rate. For malignant tumors, the primary objective was to achieve favorable postoperative gait as much as possible, not relying completely on total resection, but for benign tumors (ependymoma, hemangioma, and hemangioblastoma), total resection is desirable and should be attempted.3,6,7,14,24,28 Maintaining total resection may aggravate postoperative neurological manifestations, leading to poor gait (McCormick Scale Grades III, IV, and V), which markedly affects the quality of life of the patient. We consider that it is better to proceed with surgery while closely monitoring changes in the MEP waveform and abandon tumor resection when the waveform changes, especially when it disappears, and reengage tumor resection after postoperative improvement of the neurological symptoms.6,7,17,23,28 This study demonstrated that when walking was normal or possible without support before surgery (Grades I, II, and III), comparable gait could be maintained in 80% of the cases, and total resection was also possible in 70%. Better final outcome of walking ability was achieved in cases with good preoperative gait, and total resection was also possible in cases at an early stage with less spinal cord damage, in which gait was normal or slightly disturbed and did not require support. In such cases, the postoperative gait was less aggravated, indicating that it is desirable to perform surgery as early as possible.
Conclusions
We performed a follow-up study of 106 patients who underwent IMSCT surgery to examine outcomes, with a focus on the effect of preoperative neurological condition on postoperative mobility, and to determine the optimal timing for surgery. Among 71 patients with normal ambulation or independence without external aid (Grade I or II), 54 (76%) maintained the same mobility after surgery. Total excision was possible in 79% of patients with preoperative Grade I status. Disclaimer The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. References 1. Brotchi J: Intrinsic spinal cord tumor resection. Neurosurgery 50:1059–1063, 2002 2. Calancie B, Harris W, Brindle GF, Green BA, Landy HJ: Threshold-level repetitive transcranial electrical stimulation for intraoperative monitoring of central motor conduction. J Neurosurg 95:161–168, 2001 3. Constantini S, Miller DC, Allen JC, Rorke LB, Freed D, Epstein FJ: Radical excision of intramedullary spinal cord tu-
J. Neurosurg.: Spine / Volume 10 / May 2009
Surgical results of intramedullary spinal cord tumor mors: surgical morbidity and long-term follow-up evaluation in 164 children and young adults. J Neurosurg (2 Suppl) 93: 183–193, 2000 4. Cooper PR, Epstein F: Radical resection of intramedullary spinal cord tumors in adults. Recent experience in 29 patients. J Neurosurg 63:492–499, 1985 5. Cristante L, Herrmann HD: Surgical management of intramedullary spinal cord tumors: functional outcome and sources of morbidity. Neurosurgery 35:69–76, 1994 6. Epstein F: Spinal cord astrocytomas of childhood. Adv Tech Stand Neurosurg 13:135–169, 1986 7. Epstein FJ, Farmer JP, Freed D: Adult intramedullary spinal cord ependymomas: the result of surgery in 38 patients. J Neurosurg 79:204–209, 1993 8. Goh KY, Velasquez L, Epstein FJ: Pediatric intramedullary spinal cord tumors: is surgery alone enough? Pediatr Neurosurg 27:34–39, 1997 9. Hanbali F, Fourney DR, Marmor E, Suki D, Rhines LD, Weinberg JS, et al: Spinal cord ependymoma: radical surgical resection and outcome. Neurosurgery 51:1162–1174, 2002 10. Hoshimaru M, Koyama T, Hashimoto N, Kikuchi H: Results of microsurgical treatment for intramedullary spinal cord ependymomas: analysis of 36 cases. Neurosurgery 44:264– 269, 1999 11. Jallo GI, Freed D, Epstein F: Intramedullary spinal cord tumors in children. Childs Nerv Syst 19:641–649, 2003 12. Jallo GI, Kothbauer KF, Epstein FJ: Intrinsic spinal cord tumor resection. Neurosurgery 49:1124–1128, 2001 13. Kane PJ, el-Mahdy W, Singh A, Powell MP, Crockard HA: Spinal intradural tumors: Part II—Intramedullary. Br J Neurosurg 13:558–563, 1999 14. Kothbauer KF: Intraoperative neurophysiological monitoring for intramedullary spinal cord tumor surgery. Neurophysiol Clin 37:407–414, 2007 15. Kothbauer KF, Deletis V, Epstein FJ: Intraoperative monitoring. Pediatr Neurosurg 29:54–55, 1998 16. Kothbauer K, Deletis V, Epstein FJ: Intraoperative spinal cord monitoring for intramedullary surgery: an essential adjunct. Pediatr Neurosurg 26:247–254, 1997 17. Kothbauer KF, Deletis V, Epstein FJ: Motor-evoked potential monitoring for intramedullary spinal cord tumor surgery: correlation of clinical neurophysiological data in a series of 100 consecutive procedures. Neurosurg Focus 4:E1, 1998 18. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163–170, 1966
J. Neurosurg.: Spine / Volume 10 / May 2009
19. McCormick PC, Stein BM: Intramedullary tumors in adults. Neurosurg Clin N Am 1:609–630, 1990 20. McCormick PC, Torres R, Post KD, Stein BM: Intramedullary ependymoma of the spinal cord. J Neurosurg 72:523– 532, 1990 21. Morota N, Deletis V, Constantini S, Kofler M, Cohen H, Epstein FJ: The role of motor evoked potentials during surgery for intramedullary spinal cord tumors. Neurosurgery 41:1327–1336, 1997 22. Quiñones-Hinojosa A, Lyon R, Zada G, Lamborn KR, Gupta N, Parsa AT, et al: Changes in transcranial motor evoked potentials during intramedullary spinal cord tumor resection correlate with postoperative motor function. Neurosurgery 56:982–993, 2005 23. Sala F, Albino B, Franco F, Paola L, Massimo G: Surgery for intramedullary spinal cord tumors: the role of intraoperative (neurophysiological) monitoring. Eur Spine J 16:S130–S139, 2007 24. Sala F, Palandri G, Basso E, Lanteri P, Deletis V, Faccioli F, et al: Motor evoked potential monitoring improves outcome after surgery for intramedullary spinal cord tumors: a historical control study. Neurosurgery 58:1129–1143, 2006 25. Samii M, Klekamp J: Surgical results of 100 intramedullary tumors in relation to accompanying syringomyelia. Neurosurgery 35:865–873, 1994 26. Sandalcioglu IE, Gasser T, Wiedemayer H, Horsch S, Stolke D: Favourable outcome after biopsy and decompression of a holocord intramedullary spinal cord astrocytoma in a newborn. Eur J Paediatr Neurol 6:179–182, 2002 27. Schwartz TH, McCormick PC: Intramedullary ependymomas: clinical presentation, surgical treatment strategies and prognosis. J Neurooncol 47:211–218, 2000 28. Shrivastava RK, Epstein FJ, Perin NI, Post KD, Jallo GI: Intramedullary spinal cord tumor in patients older than 50 years of age: management and outcome analysis. J Neurosurg Spine 2:249–255, 2005
Manuscript submitted October 14, 2008. Accepted February 3, 2009. Address correspondence to: Yukihiro Matsuyama, M.D., Department of Orthopaedic Surgery, Nagoya University School of Medicine, 65 Tsuruma-Cho, Showa-Ku, Nagoya, Aichi, Japan. email:
[email protected].
413
