Eur Spine J DOI 10.1007/s00586-016-4401-z
ORIGINAL ARTICLE
Sequential alignment change of the cervical spine after anterior cervical discectomy and fusion in the lower cervical spine Ho Jin Lee1 • Doo Yong Choi1 • Myoung Hoon Shin1 • Jong Tae Kim1 Jae Taek Hong2
•
Received: 8 July 2015 / Revised: 11 January 2016 / Accepted: 15 January 2016 Ó Springer-Verlag Berlin Heidelberg 2016
Abstract Purpose The cervical spine has a linear chain of correlation or reciprocal relationship regionally (within the cervical spine) and globally (head to whole spine). The purpose of this study was to assess the sequential alignment change of the regional and global cervical spine after twolevel anterior cervical discectomy and fusion (ACDF) performed on the lower cervical spine. Methods This study included 61 patients (mean age 56 ± 8.6 years; range 35–70 years) who underwent ACDF at C5-6-7 with a plate-cage construct and whose C-spine neutral lateral radiographs showed an identical degree of horizontal gaze (occipital slope) peri-operatively. We compared the change in cervical curvature from the occiput to C7 with the absolute value (slope angle) and relative value (between two different slopes). We also investigated the correlated change in multiple angular parameters according to the change in the occipital slope. Results The occipital slope was significantly correlated with the value of the C1-slope (r = 0.33) and C2- slope (r = 0.51). The value of the center of the sellar turcica-C7 sagittal vertical axis (St-SVA) was very closely related to the C1-slope (r = -0.83), C2-slope (r = -0.8), C2-7 angle (r = -0.43), and C2-5 angle (r = -0.46). The amount of angular change at the surgical level (C5-7A) was
& Ho Jin Lee [email protected] 1
Department of Neurosurgery, Incheon St. Mary’s Hospital, The Catholic University of Korea, 56, Dongsu-ro, Bupyeong-Gu, Incheon 403-720, Republic of Korea
2
Department of Neurosurgery, St. Vincent Hospital, The Catholic University of Korea, Suwon, Republic of Korea
5.8° (2.9° -[ 8.5°), and the sum of the change in the C5slope and C7-slope was 6° (3.1° ? 2.9°). In general, the C2-5 angle decreased about 3°, in proportion to the upward inclination of C5-slope (3.1°), because the C2-slope was fixed. However, patients who showed improvement in cervical alignment (greatly increased C5-7 lordosis or greatly decreased St-SVA after surgery) often had upper cervical slope change (C1-s and C2-s). Conclusions The ACDF procedure itself can induce regional slope change (C5-s and C7-s) directly at the surgical level and can also influence upper cervical slope change (C1-s and C2s) indirectly. Then the change in the upper cervical spine can induce a change in the St-CVA and spino-cranial angle (SCA). Keywords Cervical spine curvature Anterior cervical discectomy and fusion Upper cervical spine segment
Introduction The cervical spine has a linear chain of correlation linking focally and regionally (within the cranio-cervical spine) to maintain economic balance [1]. The cervical spine can be divided into two parts: the upper cervical spine (occiput-atlas-axis; Oc-C1-C2) and lower cervical spine (C3–C7). Each part of the spine has an intimate reciprocal relationship (negative angular correlation) [2, 3]. A post-operative change in alignment usually happens in the lower cervical spine after fusion of the upper cervical spine (downward sequential change). Hyper-extended occipito-cervical (O-C) fusion often induces post-operative kyphotic deformity in the lower cervical region [4, 5]. However, upward sequential change (the upper cervical or
123
Eur Spine J
upper adjacent segment) after fusion of the lower cervical spine has not been clearly defined before. It is important to understand upward sequential change because the lower cervical spine is a common target for spine surgery. Anterior cervical discectomy and fusion (ACDF) is the most well accepted and commonly performed procedure for symptomatic cervical degenerative disease. Therefore, upward sequential change through the ACDF procedure may provide important information about regional or global change in cranio-cervical alignment and re-distribution of cervical biomechanics. The purpose of this study was to assess the sequential alignment change of the regional and global cervical spine after two-level ACDF performed on the lower cervical spine.
assessed by two neuro-spine surgeons, and the reliability of their studies (intra-class correlation coefficient score, ICCs) was assessed independently by another neuro-spine surgeon (30 patients, randomly selected). Angular measurement was conducted by the Cobb method. Figure 1 describes the measurement techniques in detail. The slope angle is positive when the specific line is oriented upwards from the horizontal line and negative when it is oriented downwards. We defined lordosis as a positive value and kyphosis as a negative value. The following parameters (linear and angular values) were measured using m-view 5.4 (Marosis Technologies, Inc., Seoul, Korea): Occipital slope (O-s)
Method We retrospectively reviewed 117 patients who underwent two-level ACDF at C5-6-7 due to symptomatic degenerative cervical disease (cervical spondylotic myelopathy or radiculopathy) between October 2007 and February 2013. To reduce the different effect of post-operative adjacent segment disease (ASD) or subsidence problems due to the surgical construct, only those patients who were given a plate-cage (allobone spacer) construct were included in this study. In general, the frequency and amount of subsidence is higher in cases with a stand-alone cage (excepting standalone anchored cage), and subsidence can induce segmental angular change during follow-up [6–8]. Nemoto et al. showed that plate-based technique induced the lower rate of segmental angular change (index-level) than standalone anchored cage technique with a 2-year follow-up [9]. Therefore, patients who underwent two-level ACDF with a stand-alone cage, stand-alone anchored cage or total disc replacement (TDR) construct were excluded from this study. Patients who had experienced trauma, neoplasm, revision operation, cervical deformity, or infection were also excluded. Because we defined the C7 vertebra as the lower margin for the cranio-cervical system, patients who showed an ambiguous low-end plate of the C7 vertebral body due to a short neck were also excluded. The height of each cage (5–8 mm) was determined according to the intraoperative measurements of the inter-vertebral disc space for improved stability. This study was approved by the institutional review board. Radiographic evaluation A C-spine neutral lateral radiograph was acquired with the patient in a comfortable, upright standing position, both arms positioned naturally at the sides of the body, preoperatively and post-operatively. These radiographs were
123
C1 (Atlas) slope (C1-s)
C2 (Axis) slope (C2-s) C5 slope (C5-s)
C7 slope (C7-s)
Occipito-C1 angle (O-C1 A) C1-2 angle (C1-2 A) Occipito-C2 angle (O-C2 A) Occiput-C7 angle (O-C7 A) Segmental C2-C5 angle (C2-5 A) Segmental C5-7 angle (C5-7 A) Global C2-C7 angle (C2-7 A) Spino-cranial angle (SCA)
The angle between the McGregor line (between the hard palate and opisthion) and horizontal line. The angle between the C1 line (between the lower margin of the anterior arch and the lower margin of the posterior arch) and horizontal line. The angle between the C2 line (lower endplate of the C2 body) and horizontal line. The angle between the C5 line (upper endplate of the C5 body) and horizontal line. The angle between the C7 line (lower endplate of the C7 body) and horizontal line. The McGregor line and C1 line were used according to Cobb’s method. The C1 and C2 lines were used according to Cobb’s method. The McGregor line and C2 line were used according to Cobb’s method. The McGregor line and C7 line were used. The C2 and C5 lines were used. The C5 and C7 lines were used. The C2 and C5 lines were used. The angle between the C7 line and the line joining the center of the sellar turcica (cST) and C7 reference point (the center of the lower endplate of the C7 body) was used.
Eur Spine J
Fig. 1 Angular measurement was conducted pre-operatively and post-operatively. The horizontal gaze is fixed. Left pre-operative state, right post-operative state. a Occipital-slope, b C1-slope, c C2-slope,
d C5-slope, e C7-slope, f spino-cranial angle(SCA), g center of the sellar turcica-C7 sagittal vertical axis (St-SVA), square center of sellar turcica, asterisk center of C7 body
cST-C7 sagittal vertical axis (St-SVA)
(31 male, 30 female; mean age 56 ± 8.6 years, range 35–70) were included in this study finally.
Translation of the cranio-cervical spine in the sagittal plane, distance between a plumb line dropped from the cST and the center of the C7 body.
Patient sub-groups Sub-groups based on the O-s values during follow-up
Patients standardization with maintaining one’s horizontal gaze The upper cervical spine is the most mobile segment of the spine and has the greatest capacity to regulate the visual axis. Therefore, we needed to establish a prior condition in the upper cervical spine to compare the change in cervical alignment. We adopted the degree of horizontal gaze as the major precondition to conduct this study because maintaining horizontal gaze is the most important functional role of the cervical spine, and the upper cervical spine plays the primary role. Since the occipital slope (O-s) is a postural variable reflecting the position of the skull, it can reflect the degree of horizontal gaze. We squared the O-s value pre-surgically and compared it with the O-s value squared after surgery to assess craniocervical alignment. Due to the possibility of measurement error, we determined the maximum difference in the O-s value before and after surgery as 2°. As a result, 61 patients
We assessed the effect of the different O-S values on other parameters. Subsidence may cause a regional change in lordosis at the C5-7 level during follow-up, which may contribute to sequential measurement error. Therefore, two individual follow-up radiographs (after surgery) were selected, each showing a different O-s value (the minimum difference value was 5°) and equal C5-7 A value (the maximum difference value was 3°). The difference in the values was calculated, and correlation analysis was performed among all parameters 37 patients). Subgroups based on the outcome at the surgical level (C5-7A) We assessed whether different radiological outcomes can induce different cervical alignment changes. We classified all patients into two subgroups according to the amount of the C5-7A change: group A, the value of C5-7A increased
123
Eur Spine J
by more than 6°; group B, the value of C5-7A increased by less than 6° or decreased. Subgroups based on the St-SVA change We classified all patients into three subgroups according to the amount of change in the St-SVA change: group D (decreased), the St-SVA value decreased by more than 5 mm; group S (similar), the St-SVA value decreased or increased by less than 5 mm; group I (increased), the StSVA value increased by more than 5 mm. Statistical analysis Study data were analyzed with descriptive statistics, frequency analysis, Chi-square test, paired t tests, and analysis of variance (ANOVA). Scheffe’s multiple comparison test was used to assess statistical significance. Differences in the study variables in relation to age, follow-up duration, or sex were analyzed with unpaired t tests or Wilcoxon signed rank tests. Pearson correlations were performed with SAS V.9.3 software. Statistical significance was set at P \ 0.05.
Results The ICCs were very high or almost perfect for all parameters (range 0.803–0.998). The pre-operative and follow-up (F/U) data for all parameters are presented in Table 1. The mean value of the O-s was maintained as 14° peri-operatively, with significant change only for the lower cervical parameters. The amount of angular change at the surgical level (C5-7A) was 5.8° (2.9°–8.5°), and the sum of the
Table 1 The pre-operative and follow-up (F/U) data for all parameters (61 patients)
Pre (OP)
change values for C7-s and C5-s was 6° (2.9°?3.1°). The upper adjacent segment (C2-5A) showed decreased lordotic curvature of about 3°. The St-SVA and SCA values were also significantly different after surgery (P \ 0.01). Sub-groups based on the O-s values during follow-up Table 2 summarizes the results of the correlation analyses among multiple variables. The vertical gaze (O-s) was significantly correlated with the value of the C1 (r = 0.33) and C2 slope (r = 0.51). The St-SVA value was very closely correlated with the C1-s (r = -0.83), C2-s (r = -0.8), C2-7A (r = -0.43), and C2-5A (r = -0.46) values. Upper cervical spine curvature (O-C2A) showed an inverse relationship with lower cervical spine curvature (C2-7A, r = -0.54 and C2-5A, r = -0.53). The SCA was strongly correlated with the C7-s parameter and O-s (r = 0.38). Subgroups based on the outcome at the surgical level (C5-7A) There was no significant difference in age, gender, or duration of follow-up between groups A and B (Table 3). Group A showed more kyphotic curvature than group B at the surgical level (C5-7A: -0.3° vs 6.1°) and more lordotic curvature at the upper cervical spine (O-C2A: 22.8° vs 17.9°) pre-operatively. Improvement in regional lordosis at the C5-7 level (group A: -0.3° -[ 10.5°) induced improvement in global spine curvature (O-C7) through the improvement at C2-7A. However, the value of C2-5A decreased after surgery (10.3° -[ 5.9°). After the ACDF
Post (OP)
D Post (OP) - pre (OP)
O-C7A (°)
34.3 ± 9.7
37.7 ± 7.9
3.3 ± 7.4
\0.01
C2-7A (°)
14.3 ± 12.5
17.1 ± 10.3
2.8 ± 10
\0.05
C2-5A (°)
11.5 ± 8.4
C5-7A (°)
2.9 ± 7.5
8.5 ± 8
-3 ± 8
\0.01
8.6 ± 6.7
5.8 ± 6.9
\0.001
O-C2A (°)
20.4 ± 9.3
20.6 ± 8.2
O-C1A (°)
-8.1 ± 7.7
-8.7 ± 7
C1-2 A (°)
28.5 ± 5
29.3 ± 5.3
St-SVA (mm)
20.7 ± 16.9
18.1 ± 14.6
0.2 ± 6
0.81
-0.6 ± 5.2
0.25
0.7 ± 3.6
0.11
-2.7 ± 9.5
\0.05
C7-s (°)
-20.9 ± 8.5
-23.8 ± 6.9
-2.9 ± 7.1
SCA (°)
103.1 ± 8.9
106.9 ± 6.8
3.8 ± 6.6
O-s (°)
\0.01 \0.001
14 ± 8.4
14 ± 5.3
0 ± 1.1
0.97
C1-s (°)
22.4 ± 8.3
23.1 ± 7.9
0.8 ± 5.6
0.07
C2-s (°) C5-s (°)
-6.4 ± 9.6 -18.1 ± 7.4
-6.4 ± 8.6 -14.9 ± 7
0 ± 6.4 3.1 ± 6.6
0.98 \0.01
Mean value ± SD, pre (OP) pre-operative, post (OP) post-operative, D difference value
123
P value
Eur Spine J Table 2 Correlation analysis of multiple variables with varying horizontal gaze values (occipital slope) during the follow-up period St-SVA St-SVA
O-C7 A
C2-7 A
C5-7 A
O-C2 A
0.15
C2-7 A
-0.43a
0.57a
C2-5 A
a
a
0.86a
a
0.58
0.64a
-0.46
0.35
-0.14
SCA
O-s
C1-s
C2-s
1 0.16
1
0.30
-0.54a
-0.53a
-0.25
1
O-C1A
0.62a
0.36a
-0.36a
-0.31
-0.23
0.85a
-0.16
0.66a
C1-2 A
a
0.30
0.09
-0.46
-0.48
-0.37a
-0.81a
-0.50a
-0.26
a
a
-0.25
0.78 a
O-s C1-s
-0.36 -0.83a
C2-s
-0.8a
a
C7-s
1
0.64a
SCA
C1-2 A
1
O-C2 A
C7-s
O-C1A
1
O-C7 A
C5-7 A
C2-5 A
a
0.78
0.56
a
1 0.24
1
-0.57a
-0.29
-0.28
-0.11
a
-0.04
0
-0.06
-0.75a
a
0.66
a
1 1
0.36 -0.11
0.23 0.49a
0.35 0.51a
-0.08 0.18
0.23 -0.69a
0.31 -0.79a
0.12 -0.12
0 0.27
0.38a 0.25
1 0.33a
1
0.04
0.65a
0.72a
0.17
-0.63a
-0.46a
-0.45a
0.19
0.36a
0.51a
0.76a
1
Statistical significance
Table 3 Sequential change in cervical alignment after ACDF (subgroups indicate surgical level outcomes: lordotic change of C5-C7) Group
A (n = 31) Pre-op
A (n = 31) Post-op
B (n = 30) Pre-op
B (n = 30) Post-op
A vs B (pre-OP) P value
Age (years)
55.6 ± 8.2
54.4 ± 8.3
0.57
Sex (M/F)
(14/17)
(17/13)
0.73
F/U (months)
12.2 ± 11.6
14.7 ± 15.8
0.29
a
O-C7A (°)
32.1 ± 11
39 ± 8.2
36.8 ± 7.6
36.3 ± 7.5
C2-7A (°)
10 ± 11.3
16.5 ± .9.7a
18.8 ± 12.2
17.8 ± 11.1
C2-5A (°)
10.3 ± 9.3
5.9 ± 9a
12.7 ± 7.4
C5-7A (°) O-C2A (°)
-0.3 ± 6.1 22.8 ± 9.1
10.5 ± 5a 22.5 ± .8.2
O-C1A (°)
-6.1 ± 7.3
C1-2 A (°)
29 ± 4.4
0.58
A vs B (post-OP) P value
0.19
\0.01
0.64
11.1 ± 5.9
0.26
\0.05
6.1 ± 7.4 17.9 ± 8.9
6.7 ± 7.7 18.6 ± 7.9
\0.001 \0.05
\0.01 0.06
-7 ± 6.9
-10.2 ± 7.6
-10.4 ± 6.9
\0.05
0.06
29.6 ± 4.8
28.1 ± 5.5
29 ± 5.8
0.50
0.55 0.05
St-SVA (mm)
25.3 ± 15
21.6 ± 12.7
16 ± 17.5
14.4 ± 15.8
\0.05
C7-s (°)
-19.4 ± 9.9
-25 ± 7.2a
-22.4 ± 6.5
-22.5 ± 6.5
0.17
0.30
SCA (°)
100.5 ± 8.7
106.9 ± 6.7a
105.8 ± 8.5
106.8 ± 7.1
\0.05
0.64
O-s (°)
13.6 ± 5.5
13.7 ± 5.2
14.5 ± 5.4
14.3 ± 5.4
0.51
0.64
C1-s (°)
19.6 ± 7.7
21.2 ± 7.7
25.1 ± 8
25.2 ± 7.7
\0.05
\0.05
C2-s (°)
-9.2 ± 8.9
-8.5 ± 9
-3.4 ± 9.5
-4.3 ± 7.7
\0.05
\0.05
C5-s (°)
-19.9 ± 8.9
-14.5 ± 8.2a
-16.2 ± 6.6
-15.4 ± 5.5
\0.05
0.35
a
Statistical significant change after surgery, pre-op pre-operative, post-op post-operative, A (group A) the value of C5-7A increased by more than 6°, B (group B) the value of C5-7A increased by less than 6° or decreased
operation, the C5-s showed upward inclination (5.4°), and C7-s showed downward inclination (5.6°). The summation of change values of the C5 and C7 slope (11°) was nearly the same as the regional change at the C5-7 level (10.8°). Patients in group B (those who maintained similar regional curvature at the C5-7 level) did not show any change in parameters post-operatively. Groups A and B had similar values for the upper cervical spine parameters (O-C2A, O-C1A) peri-operatively, when the O-slope was fixed.
Subgroups based on the St-SVA change (Table 4) There was no significant difference in age, gender, or duration of follow-up among the three groups perioperatively. Group D showed more kyphotic curvature segmentally (C5-7A, C2-5A) and globally (C2-7A) at the lower cervical spine and more lordotic curvature at the upper cervical spine (O-C2A; 25.8° vs 18.7° vs 12.9°) than groups S and I pre-operatively. The St-CVA value in group
123
123
-12.3 ± 7.9
-20.1 ± 7.4
C2-s (°)
C5-s (°)
14.5 ± 4.9 24.5 ± 7.8 -4.2 ± 8.7 -16.9 ± 7.1
-7.8 ± 8.8a -14.8 ± 7.2a
104.8 ± 6.4
-20.9 ± 5.4
13.9 ± 6.1
a
28.7 ± 4.8 16.1 ± 16.1
22.4 ± 7.1a
-23 ± 8.4
30.3 ± 4.9 18.3 ± 10.9a
a
14.4 ± 4.4
-15 ± 7.1
-5.3 ± 8.2
24.9 ± 7.3
-16.5 ± 7.6
1.2 ± 7.6
27.3 ± 7.4
14.1 ± 5.3
108 ± 9.2
-23.7 ± 8.2
-24 ± 6.1a 108.1 ± 5.4a
25.3 ± 5.8 12.2 ± 17.8
-12.4 ± 4.6
12.9 ± 6.3
7.2 ± 8.4
17.7 ± 7.2
24.9 ± 11.1
37.5 ± 10.4
11.7 ± 9.3
(7/4)
55.5 ± 8.3
I (Pre-OP) (n = 11)
29.7 ± 5.1 15.5 ± 15.3
-10 ± 6.5
19.7 ± 7.1
9.2 ± 6.1
9.7 ± 5.9
18.9 ± 8.2
38.3 ± 6.4a
S (post-OP)
-15.1 ± 6.9
-6.1 ± 9.6a
20.5 ± 10.4a
13.4 ± 5.7
106.5 ± 8.7
-24.8 ± 5.4
26.2 ± 5.8 23.6 ± 19.5a
-6.7 ± 6.3a
19.5 ± 7.6a
0.23
\0.01
\0.01
0.83
0.051
0.40
0.68 \0.05
\0.05
\0.05
0.17
\0.05
9 ± 6.6a 9.3 ± 6.9
\0.05
0.31
0.41
0.52
0.60
D vs S (pre-OP) P value
18.3 ± 11.3a
37.7 ± 8.3
I (Post-OP)
\0.001
\0.01
\0.05
0.20
0.58
0.57
0.65 0.14 0.77
\0.01
0.15
0.40
0.19
0.10
S vs I (pre-OP) P value
\0.05 \0.05
\0.001
\0.05
\0.01
\0.001
D vs W (pre-OP) P value
0.99
0.59
0.25
0.86
0.47
0.75
0.09 0.31
0.37
0.49
0.68
0.46
0.33
0.81
D vs S vs I (post -OP) P value
Statistically significant change after surgery, pre-op pre-operative, post-op post-operative, D (decreased group) the St-SVA value decreased by more than 5 mm, S (similar group) the St-SVA value decreased or increased by less than 5 mm, I (increased group) the St-SVA value increased by more than 5 mm
a
13.5 ± 6.1
17.8 ± 7.1
99 ± 9.7
SCA (°)
O-s (°)
-19.5 ± 11.6
C7-s (°)
C1-s (°)
105.8 ± 7.3
29.9 ± 4.3 29.7 ± 13.4
C1-2 A (°) St-SVA (mm)
-10 ± 6.4
-8.1 ± 7.9a
-4.1 ± 8.4
3.7 ± 6.3
12.7 ± 7.1
16.4 ± 10.3
35 ± 6.9
18.7 ± 7.7
a
11.3 ± 10
(11/15)
56 ± 8.7
S (pre-OP) (n = 26)
22.2 ± 9.6a
7.7 ± 7.3
a
6.9 ± 10.3
O-C1A (°)
C2-5A (°)
14.7 ± 11.9
-0.1 ± 7.2
7.3 ± 8.4
C2-7A (°)
36.9 ± 9.4a
25.8 ± 9.1
7.3 ± 11.2
O-C7A (°)
O-C2A (°)
32.3 ± 11.6
F/U (months)
D (post-OP)
C5-7A (°)
(13/11)
16.3 ± 18.1
Sex (M/F)
53.7 ± 7.8
Age (years)
D (pre-OP) (n = 24)
Table 4 Sequential change in cervical alignment according to St-SVA value outcomes (I: increased, S: similar, D: decreased)
Eur Spine J
Eur Spine J
D was also significantly larger than that of groups S and I pre-operatively. The curvature at the surgical level (C5-7A) was not significantly different (-0.1° vs 3.7°) between groups D and S; however, the upper adjacent level (C2-5A) was significantly lower in group D (7.3°) than group S (12.7°) pre-operatively. When comparing groups D and I, the curvature of both the surgical (C5-7) and upper adjacent level (C2-5) was also significantly different (P \ 0.05) pre-operatively. After ACDF, the C1 and C2 slopes of group D showed an upward inclination (4.5° and 4.6°), which was similar to the amount of C5-s change (5.3°). However, the inclination of the C1-s and C2-S of other groups (group S and I) was maintained or showed post-operative downward movement. After ACDF, none of the parameters in the three groups were significantly different regardless of the preoperative alignment of the cervical spine.
Discussion Horizontal gaze has a significant impact on activities of daily living and quality of life, and it can reflect the position of the skull [10]. In the present study, we adopted the McGregor line (O-s) as a substitute for the Frankfort horizontal (FH) line (the imaginary line between the lower border of the orbit and the external auditory canal) [11]. Although the Frankfort horizontal line is the representative cephalometric reference line to determine the degree of vertical gaze, we found that the anatomical margin of the external auditory canal (EAC) could be more ambiguous than the sellar turcica. Moreover, the McGregor line is also a good postural variable reflecting the position of the skull, and it can indicate the degree of vertical alteration in a person’s gaze. We also found a close relationship between the FH line and McGregor line in this study. If we fixed the O-s value peri-operatively (30 patients, randomly selected), the pre-operative value of the FH line slope remained similar after surgery (mean value, before surgery: 20.3°; after surgery: 20.4°). In general, sequential change in cranio-cervical alignment after ACDF appears to follow the pattern described below when the horizontal gaze is fixed (Table 1; Fig. 2). Insertion of the interbody graft results in vertical distractive forces to the surgical segment (C5-6-7), which induces upward movement of C5-s and downward movement of C7-s (D 2.9°). An upward movement of C5-s (D 3.2°) induces a decrease in C2-5A (D 3°) because C2-s does not change. A downward movement of C7-s determines the C7 reference point to a forward direction, which can induce an increase in the SCA (D 3.8°). C7-s was mathematically related to SCA (r = -0.75), and Le Huec has previously
described the formula: SCA = 90° - C7 s ? sellar turcica tilt [1]. Peri-operatively, none of the parameters related to the upper cervical region (O-C2A, C1-2A, C1-s and C2-s) changed when the horizontal gaze was fixed. This result may provide insight on the likelihood of upper cervical movement. First, maintaining horizontal gaze was not controlled by the O-C1 joint separately; rather, its function expanded to the C1-2 joint. Therefore, the O-C1-C2 segments are closely connected and act as a unified segment. Second, the degree of horizontal gaze can reflect upper cervical spine alignment (O-C2). This phenomenon is affirmed by the correlation between sagittal spinal parameters. Since the O-s was closely related to the C2-s (r = 0.51) and C1-S (r = 0.33), the difference in horizontal gaze can indicate the difference in the values of C1-s and C2-s. An upward horizontal gaze induced the upward movement of C1-s and C2-S concurrently, which may have induced backward movement of the center of gravity (sellar turcica), resulting in a decrease in the St- SVA (r = -0.36) value. However, parameters related to the surgical level (C5-7A, C7-s) were not significantly correlated with the change in the O-s value (r = -0.08, r = 0, respectively). After ACDF, the change in horizontal gaze was closely correlated with the upper adjacent segment (C2-5) and upper cervical segment (O-C2) rather than the surgical level (C5-7). To assess radiological outcomes after ACDF, we determined the degree of angle correction (D C5-7A) at the surgical level. Because the values of O-s and C7-s were similar before surgery, the pre-operative O-C7A value did not differ between the two groups (A and B) (P = 0.58). When analyzing the C2-7A as C2-5A and C5-7A, only C57A differed significantly between the two groups before surgery (P \ 0.001). After ACDF, the C2-7A of group A was similar to that of group B due to the improvement in C5-7A lordosis (D 10.8°). Sequentially, the C2-5A value of group A decreased (D 4.4°) in proportion to the upward inclination of C5-s (D 5.4°) and C2 (D 0.7°). Therefore, the ACDF procedure induced sequential change in cervical alignment mainly at the lower cervical spine (C2-C7) in group A, which showed segmental kyphosis (C5-7A) preoperatively. Patients in group B who had sufficient segmental lordosis at the surgical level (C5-7) and the upper adjacent level (C2-5) before surgery showed no significant differences perioperatively (P [ 0.05). Nonsignificant upward inclination of the C1-S (D 1.6°) and C2-s (D 0.7°) in group A and downward inclination of the C2-S (D 0.9°) in group B were found. We assumed that the St-SVA was another representative factor for estimating radiological outcome because it is well known that the degree of St-SVA is closely related to
123
Eur Spine J
Fig. 2 General sequential change in cranio-cervical alignment after ACDF, when the horizontal gaze is fixed. a Occipital-slope, b C5-slope, c C7-slope, d spino-cranial angle (SCA), e center of the sellar turcica-C7 sagittal vertical axis (St-SVA), - decrement, ? increment
the clinical outcome in terms of neck pain and healthrelated quality of life (HRQOL) [12, 13]. In general, preoperative spinal curvature differed among the three groups (D, S, and I), with the exception of O-C7A and C7-S. Group D had more kyphotic curvature than groups S and I at the lower cervical spine pre-operatively, and the St-SVA value was also greater in group D than in the other two groups. After ACDF, group D showed restored regional lordosis at the lower cervical spine (C5-7A: D 7.8°, C2-5A: D 1.6°) and reduced lordosis at the upper cervical spine (OC2A: 25.8° -[ 22.2°). The upward inclination of C1-s and C2-s induced a decrease in O-C2A. Therefore, the ACDF procedure in group D induced sequential change in cervical curvature that expanded to the upper cervical spine (O-C2), resulting in a significant decrease in the St-SVA value (Fig. 3). On the other hand, group S and I also showed restored segmental lordosis at the surgical level (C5-7A: D 5.5° and D 2.1°); however, the inclination of C1-s and C2-S was maintained similarly (group S) or showed a downward movement (group I) after surgery. Therefore, St-CVA was either unchanged (group S) or aggravated (group I). These results show that the pre-operative state of C2-5A (upper adjacent level), C1-s, and C2-s may be the key determinant of improvement in St-SVA. In other words, the
123
state of degeneration in the cervical spine may determine the radiological outcome after ACDF. If one patient (group D, relatively high degenerative state) shows regional kyphosis at the lower cervical spine (C2-5 and C5-7) and higher lordosis at the upper cervical spine (O-C2) before surgery, ACDF can restore regional lordosis at the lower cervical spine (C2-7A), which then induces kyphosis at the upper cervical spine. If another patient (group I, relatively non-degenerative) shows the opposite state of cervical curvature, ACDF can decrease regional lordosis at the lower cervical spine (C2-7A), which then induces lordotic change at the upper cervical spine. Ironically, although segmental lordosis at the surgical level (C5-7) was improved in group I, regional lordosis at C2-7A decreased after ACDF. Such a finding results from the downward movement of C2-s. We did not determine the exact physiologic origins of cervical spine changes after ACDF, and thus there may be additional undisclosed factors. Previous reports showed that preservation or secure more lordotic curvature at surgical level (Villavicencio et al.) and decreases in the SVA-value (Tang et al.) are the most important factors for achieving good clinical outcomes related to neck pain or HRQOL after cervical spine
Eur Spine J
Fig. 3 ACDF procedure in group D (St-SVA, decreased) induced sequential change in cervical curvature that expanded to the upper cervical spine. a Occipital-slope, b C5-slope, c C7-slope, d spino-
cranial angle (SCA), e center of the sellar turcica-C7 sagittal vertical axis (St-SVA), - decrement, ? increment
surgery [12, 13]. Therefore, the present study contributes to the theoretical basis of our understanding of neck pain and HRQOL after ACDF considering upward sequential change of the cervical spine.
alignment changes was not discussed nor was the influence of minor residual stenosis at C4-5 in a patient who underwent fusion at C5-7.
Conclusion Limitation and interpretation This study was retrospective, so it was not possible to control the cervical spine position in detail. Second, the investigation of sequential change of the cervical spine was confined to the specific surgical level (C5-6-7). Future studies of anterior cervical surgery with another lesion (C45-6), multi-levels (more than three level) and more extensive corrective surgery (corpectomy) would be informative [10]. Moreover, comparative study between plate-cage system and other surgical constructs (zero-profile with selflocking system or TDR) may be also interesting [8, 9, 14]. Third, the present study did not include the thoraco-lumbar region for assessment of the effect of trunk alignment on cervical spine alignment, although consecutive patients did not show sagittal imbalance pre-operatively in this study. Fourth, the impact of pain or residual disability on
Since the regional or global cervical angle is a relative value made by two different slope lines, determining the degree of the slope angle is the most crucial step for assessing a change in spinal alignment. ACDF can induce improvement in regional lordosis at the surgical level and can subsequently influence changes in the upper cervical segment through upward inclination of C1-s and C2-s according to pre-operative kyphotic change of the lower cervical spine. Then, sequential change in cervical spine alignment can influence the StCVA and SCA values. Compliance with ethical standards Conflicts of interest All authors certify that that they have no affiliations with or involvement in any organization or entity with any financial or non-financial interest in the subject of this paper.
123
Eur Spine J
References 1. Le Huec JC, Demezon H, Aunoble S (2015) Sagittal parameters of global cervical balance using EOS imaging: normative values from a prospective cohort of asymptomatic volunteers. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc 24:63–71. doi:10.1007/s00586-014-3632-0 2. Nojiri K, Matsumoto M, Chiba K, Maruiwa H, Nakamura M, Nishizawa T, Toyama Y (2003) Relationship between alignment of upper and lower cervical spine in asymptomatic individuals. J Neurosurg 99:80–83 3. Lee SH, Kim KT, Seo EM, Suk KS, Kwack YH, Son ES (2012) The influence of thoracic inlet alignment on the craniocervical sagittal balance in asymptomatic adults. J Spinal Disord Tech 25:E41–E47. doi:10.1097/BSD.0b013e3182396301 4. Toyama Y, Matsumoto M, Chiba K, Asazuma T, Suzuki N, Fujimura Y, Hirabayashi K (1994) Realignment of postoperative cervical kyphosis in children by vertebral remodeling. Spine 19:2565–2570 5. Matsunaga S, Onishi T, Sakou T (2001) Significance of occipitoaxial angle in subaxial lesion after occipitocervical fusion. Spine 26:161–165 6. Lee CH, Kim KJ, Hyun SJ, Yeom JS, Jahng TA, Kim HJ (2015) Subsidence as of 12 months after single-level anterior cervical inter-body fusion. Is it related to clinical outcomes? Acta Neurochirurgica. doi:10.1007/s00701-015-2388-6 7. Lee YS, Kim YB, Park SW (2014) Risk factors for postoperative subsidence of single-level anterior cervical discectomy and fusion: the significance of the preoperative cervical alignment. Spine 39:1280–1287. doi:10.1097/brs.0000000000000400 8. Grasso G, Giambartino F, Tomasello G, Iacopino G (2014) Anterior cervical discectomy and fusion with ROI-C peek cage: cervical alignment and patient outcomes. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc 23(Suppl 6):650–657. doi:10.1007/s00586-014-3553-y
123
9. Nemoto O, Kitada A, Naitou S, Tachibana A, Ito Y, Fujikawa A (2015) Stand-alone anchored cage versus cage with plating for single-level anterior cervical discectomy and fusion: a prospective, randomized, controlled study with a 2-year follow-up. Eur J Orthop Surg Traumatol Orthopedie Traumatologie 25(Suppl 1):S127–S134. doi:10.1007/s00590-014-1547-4 10. Mattei TA, Rehman AA, Issawi A, Fassett DR (2015) Surgical challenges in the management of cervical kyphotic deformity in patients with severe osteoporosis: an illustrative case of a patient with Hajdu-Cheney syndrome. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. doi:10. 1007/s00586-015-4092-x 11. Lundstrom A, Lundstrom F (1995) The Frankfort horizontal as a basis for cephalometric analysis. Am J Orthodont Dentofac Orthop Off Publ Am Assoc Orthodont Const Soc Am Board Orthodont 107:537–540 12. Tang JA, Scheer JK, Smith JS, Deviren V, Bess S, Hart RA, Lafage V, Shaffrey CI, Schwab F, Ames CP (2015) The impact of standing regional cervical sagittal alignment on outcomes in posterior cervical fusion surgery. Neurosurgery 76(Suppl 1):S14– S21. doi:10.1227/01.neu.0000462074.66077.2b (discussion S21) 13. Villavicencio AT, Babuska JM, Ashton A, Busch E, Roeca C, Nelson EL, Mason A, Burneikiene S (2011) Prospective, randomized, double-blind clinical study evaluating the correlation of clinical outcomes and cervical sagittal alignment. Neurosurgery 68:1309–1316. doi:10.1227/NEU.0b013e31820b51f3 (discussion 1316) 14. Gao Y, Liu M, Li T, Huang F, Tang T, Xiang Z (2013) A metaanalysis comparing the results of cervical disc arthroplasty with anterior cervical discectomy and fusion (ACDF) for the treatment of symptomatic cervical disc disease. J Bone Jt Surg Am 95:555–561. doi:10.2106/jbjs.k.00599
ORIGINAL ARTICLE
Sequential alignment change of the cervical spine after anterior cervical discectomy and fusion in the lower cervical spine Ho Jin Lee1 • Doo Yong Choi1 • Myoung Hoon Shin1 • Jong Tae Kim1 Jae Taek Hong2
•
Received: 8 July 2015 / Revised: 11 January 2016 / Accepted: 15 January 2016 Ó Springer-Verlag Berlin Heidelberg 2016
Abstract Purpose The cervical spine has a linear chain of correlation or reciprocal relationship regionally (within the cervical spine) and globally (head to whole spine). The purpose of this study was to assess the sequential alignment change of the regional and global cervical spine after twolevel anterior cervical discectomy and fusion (ACDF) performed on the lower cervical spine. Methods This study included 61 patients (mean age 56 ± 8.6 years; range 35–70 years) who underwent ACDF at C5-6-7 with a plate-cage construct and whose C-spine neutral lateral radiographs showed an identical degree of horizontal gaze (occipital slope) peri-operatively. We compared the change in cervical curvature from the occiput to C7 with the absolute value (slope angle) and relative value (between two different slopes). We also investigated the correlated change in multiple angular parameters according to the change in the occipital slope. Results The occipital slope was significantly correlated with the value of the C1-slope (r = 0.33) and C2- slope (r = 0.51). The value of the center of the sellar turcica-C7 sagittal vertical axis (St-SVA) was very closely related to the C1-slope (r = -0.83), C2-slope (r = -0.8), C2-7 angle (r = -0.43), and C2-5 angle (r = -0.46). The amount of angular change at the surgical level (C5-7A) was
& Ho Jin Lee [email protected] 1
Department of Neurosurgery, Incheon St. Mary’s Hospital, The Catholic University of Korea, 56, Dongsu-ro, Bupyeong-Gu, Incheon 403-720, Republic of Korea
2
Department of Neurosurgery, St. Vincent Hospital, The Catholic University of Korea, Suwon, Republic of Korea
5.8° (2.9° -[ 8.5°), and the sum of the change in the C5slope and C7-slope was 6° (3.1° ? 2.9°). In general, the C2-5 angle decreased about 3°, in proportion to the upward inclination of C5-slope (3.1°), because the C2-slope was fixed. However, patients who showed improvement in cervical alignment (greatly increased C5-7 lordosis or greatly decreased St-SVA after surgery) often had upper cervical slope change (C1-s and C2-s). Conclusions The ACDF procedure itself can induce regional slope change (C5-s and C7-s) directly at the surgical level and can also influence upper cervical slope change (C1-s and C2s) indirectly. Then the change in the upper cervical spine can induce a change in the St-CVA and spino-cranial angle (SCA). Keywords Cervical spine curvature Anterior cervical discectomy and fusion Upper cervical spine segment
Introduction The cervical spine has a linear chain of correlation linking focally and regionally (within the cranio-cervical spine) to maintain economic balance [1]. The cervical spine can be divided into two parts: the upper cervical spine (occiput-atlas-axis; Oc-C1-C2) and lower cervical spine (C3–C7). Each part of the spine has an intimate reciprocal relationship (negative angular correlation) [2, 3]. A post-operative change in alignment usually happens in the lower cervical spine after fusion of the upper cervical spine (downward sequential change). Hyper-extended occipito-cervical (O-C) fusion often induces post-operative kyphotic deformity in the lower cervical region [4, 5]. However, upward sequential change (the upper cervical or
123
Eur Spine J
upper adjacent segment) after fusion of the lower cervical spine has not been clearly defined before. It is important to understand upward sequential change because the lower cervical spine is a common target for spine surgery. Anterior cervical discectomy and fusion (ACDF) is the most well accepted and commonly performed procedure for symptomatic cervical degenerative disease. Therefore, upward sequential change through the ACDF procedure may provide important information about regional or global change in cranio-cervical alignment and re-distribution of cervical biomechanics. The purpose of this study was to assess the sequential alignment change of the regional and global cervical spine after two-level ACDF performed on the lower cervical spine.
assessed by two neuro-spine surgeons, and the reliability of their studies (intra-class correlation coefficient score, ICCs) was assessed independently by another neuro-spine surgeon (30 patients, randomly selected). Angular measurement was conducted by the Cobb method. Figure 1 describes the measurement techniques in detail. The slope angle is positive when the specific line is oriented upwards from the horizontal line and negative when it is oriented downwards. We defined lordosis as a positive value and kyphosis as a negative value. The following parameters (linear and angular values) were measured using m-view 5.4 (Marosis Technologies, Inc., Seoul, Korea): Occipital slope (O-s)
Method We retrospectively reviewed 117 patients who underwent two-level ACDF at C5-6-7 due to symptomatic degenerative cervical disease (cervical spondylotic myelopathy or radiculopathy) between October 2007 and February 2013. To reduce the different effect of post-operative adjacent segment disease (ASD) or subsidence problems due to the surgical construct, only those patients who were given a plate-cage (allobone spacer) construct were included in this study. In general, the frequency and amount of subsidence is higher in cases with a stand-alone cage (excepting standalone anchored cage), and subsidence can induce segmental angular change during follow-up [6–8]. Nemoto et al. showed that plate-based technique induced the lower rate of segmental angular change (index-level) than standalone anchored cage technique with a 2-year follow-up [9]. Therefore, patients who underwent two-level ACDF with a stand-alone cage, stand-alone anchored cage or total disc replacement (TDR) construct were excluded from this study. Patients who had experienced trauma, neoplasm, revision operation, cervical deformity, or infection were also excluded. Because we defined the C7 vertebra as the lower margin for the cranio-cervical system, patients who showed an ambiguous low-end plate of the C7 vertebral body due to a short neck were also excluded. The height of each cage (5–8 mm) was determined according to the intraoperative measurements of the inter-vertebral disc space for improved stability. This study was approved by the institutional review board. Radiographic evaluation A C-spine neutral lateral radiograph was acquired with the patient in a comfortable, upright standing position, both arms positioned naturally at the sides of the body, preoperatively and post-operatively. These radiographs were
123
C1 (Atlas) slope (C1-s)
C2 (Axis) slope (C2-s) C5 slope (C5-s)
C7 slope (C7-s)
Occipito-C1 angle (O-C1 A) C1-2 angle (C1-2 A) Occipito-C2 angle (O-C2 A) Occiput-C7 angle (O-C7 A) Segmental C2-C5 angle (C2-5 A) Segmental C5-7 angle (C5-7 A) Global C2-C7 angle (C2-7 A) Spino-cranial angle (SCA)
The angle between the McGregor line (between the hard palate and opisthion) and horizontal line. The angle between the C1 line (between the lower margin of the anterior arch and the lower margin of the posterior arch) and horizontal line. The angle between the C2 line (lower endplate of the C2 body) and horizontal line. The angle between the C5 line (upper endplate of the C5 body) and horizontal line. The angle between the C7 line (lower endplate of the C7 body) and horizontal line. The McGregor line and C1 line were used according to Cobb’s method. The C1 and C2 lines were used according to Cobb’s method. The McGregor line and C2 line were used according to Cobb’s method. The McGregor line and C7 line were used. The C2 and C5 lines were used. The C5 and C7 lines were used. The C2 and C5 lines were used. The angle between the C7 line and the line joining the center of the sellar turcica (cST) and C7 reference point (the center of the lower endplate of the C7 body) was used.
Eur Spine J
Fig. 1 Angular measurement was conducted pre-operatively and post-operatively. The horizontal gaze is fixed. Left pre-operative state, right post-operative state. a Occipital-slope, b C1-slope, c C2-slope,
d C5-slope, e C7-slope, f spino-cranial angle(SCA), g center of the sellar turcica-C7 sagittal vertical axis (St-SVA), square center of sellar turcica, asterisk center of C7 body
cST-C7 sagittal vertical axis (St-SVA)
(31 male, 30 female; mean age 56 ± 8.6 years, range 35–70) were included in this study finally.
Translation of the cranio-cervical spine in the sagittal plane, distance between a plumb line dropped from the cST and the center of the C7 body.
Patient sub-groups Sub-groups based on the O-s values during follow-up
Patients standardization with maintaining one’s horizontal gaze The upper cervical spine is the most mobile segment of the spine and has the greatest capacity to regulate the visual axis. Therefore, we needed to establish a prior condition in the upper cervical spine to compare the change in cervical alignment. We adopted the degree of horizontal gaze as the major precondition to conduct this study because maintaining horizontal gaze is the most important functional role of the cervical spine, and the upper cervical spine plays the primary role. Since the occipital slope (O-s) is a postural variable reflecting the position of the skull, it can reflect the degree of horizontal gaze. We squared the O-s value pre-surgically and compared it with the O-s value squared after surgery to assess craniocervical alignment. Due to the possibility of measurement error, we determined the maximum difference in the O-s value before and after surgery as 2°. As a result, 61 patients
We assessed the effect of the different O-S values on other parameters. Subsidence may cause a regional change in lordosis at the C5-7 level during follow-up, which may contribute to sequential measurement error. Therefore, two individual follow-up radiographs (after surgery) were selected, each showing a different O-s value (the minimum difference value was 5°) and equal C5-7 A value (the maximum difference value was 3°). The difference in the values was calculated, and correlation analysis was performed among all parameters 37 patients). Subgroups based on the outcome at the surgical level (C5-7A) We assessed whether different radiological outcomes can induce different cervical alignment changes. We classified all patients into two subgroups according to the amount of the C5-7A change: group A, the value of C5-7A increased
123
Eur Spine J
by more than 6°; group B, the value of C5-7A increased by less than 6° or decreased. Subgroups based on the St-SVA change We classified all patients into three subgroups according to the amount of change in the St-SVA change: group D (decreased), the St-SVA value decreased by more than 5 mm; group S (similar), the St-SVA value decreased or increased by less than 5 mm; group I (increased), the StSVA value increased by more than 5 mm. Statistical analysis Study data were analyzed with descriptive statistics, frequency analysis, Chi-square test, paired t tests, and analysis of variance (ANOVA). Scheffe’s multiple comparison test was used to assess statistical significance. Differences in the study variables in relation to age, follow-up duration, or sex were analyzed with unpaired t tests or Wilcoxon signed rank tests. Pearson correlations were performed with SAS V.9.3 software. Statistical significance was set at P \ 0.05.
Results The ICCs were very high or almost perfect for all parameters (range 0.803–0.998). The pre-operative and follow-up (F/U) data for all parameters are presented in Table 1. The mean value of the O-s was maintained as 14° peri-operatively, with significant change only for the lower cervical parameters. The amount of angular change at the surgical level (C5-7A) was 5.8° (2.9°–8.5°), and the sum of the
Table 1 The pre-operative and follow-up (F/U) data for all parameters (61 patients)
Pre (OP)
change values for C7-s and C5-s was 6° (2.9°?3.1°). The upper adjacent segment (C2-5A) showed decreased lordotic curvature of about 3°. The St-SVA and SCA values were also significantly different after surgery (P \ 0.01). Sub-groups based on the O-s values during follow-up Table 2 summarizes the results of the correlation analyses among multiple variables. The vertical gaze (O-s) was significantly correlated with the value of the C1 (r = 0.33) and C2 slope (r = 0.51). The St-SVA value was very closely correlated with the C1-s (r = -0.83), C2-s (r = -0.8), C2-7A (r = -0.43), and C2-5A (r = -0.46) values. Upper cervical spine curvature (O-C2A) showed an inverse relationship with lower cervical spine curvature (C2-7A, r = -0.54 and C2-5A, r = -0.53). The SCA was strongly correlated with the C7-s parameter and O-s (r = 0.38). Subgroups based on the outcome at the surgical level (C5-7A) There was no significant difference in age, gender, or duration of follow-up between groups A and B (Table 3). Group A showed more kyphotic curvature than group B at the surgical level (C5-7A: -0.3° vs 6.1°) and more lordotic curvature at the upper cervical spine (O-C2A: 22.8° vs 17.9°) pre-operatively. Improvement in regional lordosis at the C5-7 level (group A: -0.3° -[ 10.5°) induced improvement in global spine curvature (O-C7) through the improvement at C2-7A. However, the value of C2-5A decreased after surgery (10.3° -[ 5.9°). After the ACDF
Post (OP)
D Post (OP) - pre (OP)
O-C7A (°)
34.3 ± 9.7
37.7 ± 7.9
3.3 ± 7.4
\0.01
C2-7A (°)
14.3 ± 12.5
17.1 ± 10.3
2.8 ± 10
\0.05
C2-5A (°)
11.5 ± 8.4
C5-7A (°)
2.9 ± 7.5
8.5 ± 8
-3 ± 8
\0.01
8.6 ± 6.7
5.8 ± 6.9
\0.001
O-C2A (°)
20.4 ± 9.3
20.6 ± 8.2
O-C1A (°)
-8.1 ± 7.7
-8.7 ± 7
C1-2 A (°)
28.5 ± 5
29.3 ± 5.3
St-SVA (mm)
20.7 ± 16.9
18.1 ± 14.6
0.2 ± 6
0.81
-0.6 ± 5.2
0.25
0.7 ± 3.6
0.11
-2.7 ± 9.5
\0.05
C7-s (°)
-20.9 ± 8.5
-23.8 ± 6.9
-2.9 ± 7.1
SCA (°)
103.1 ± 8.9
106.9 ± 6.8
3.8 ± 6.6
O-s (°)
\0.01 \0.001
14 ± 8.4
14 ± 5.3
0 ± 1.1
0.97
C1-s (°)
22.4 ± 8.3
23.1 ± 7.9
0.8 ± 5.6
0.07
C2-s (°) C5-s (°)
-6.4 ± 9.6 -18.1 ± 7.4
-6.4 ± 8.6 -14.9 ± 7
0 ± 6.4 3.1 ± 6.6
0.98 \0.01
Mean value ± SD, pre (OP) pre-operative, post (OP) post-operative, D difference value
123
P value
Eur Spine J Table 2 Correlation analysis of multiple variables with varying horizontal gaze values (occipital slope) during the follow-up period St-SVA St-SVA
O-C7 A
C2-7 A
C5-7 A
O-C2 A
0.15
C2-7 A
-0.43a
0.57a
C2-5 A
a
a
0.86a
a
0.58
0.64a
-0.46
0.35
-0.14
SCA
O-s
C1-s
C2-s
1 0.16
1
0.30
-0.54a
-0.53a
-0.25
1
O-C1A
0.62a
0.36a
-0.36a
-0.31
-0.23
0.85a
-0.16
0.66a
C1-2 A
a
0.30
0.09
-0.46
-0.48
-0.37a
-0.81a
-0.50a
-0.26
a
a
-0.25
0.78 a
O-s C1-s
-0.36 -0.83a
C2-s
-0.8a
a
C7-s
1
0.64a
SCA
C1-2 A
1
O-C2 A
C7-s
O-C1A
1
O-C7 A
C5-7 A
C2-5 A
a
0.78
0.56
a
1 0.24
1
-0.57a
-0.29
-0.28
-0.11
a
-0.04
0
-0.06
-0.75a
a
0.66
a
1 1
0.36 -0.11
0.23 0.49a
0.35 0.51a
-0.08 0.18
0.23 -0.69a
0.31 -0.79a
0.12 -0.12
0 0.27
0.38a 0.25
1 0.33a
1
0.04
0.65a
0.72a
0.17
-0.63a
-0.46a
-0.45a
0.19
0.36a
0.51a
0.76a
1
Statistical significance
Table 3 Sequential change in cervical alignment after ACDF (subgroups indicate surgical level outcomes: lordotic change of C5-C7) Group
A (n = 31) Pre-op
A (n = 31) Post-op
B (n = 30) Pre-op
B (n = 30) Post-op
A vs B (pre-OP) P value
Age (years)
55.6 ± 8.2
54.4 ± 8.3
0.57
Sex (M/F)
(14/17)
(17/13)
0.73
F/U (months)
12.2 ± 11.6
14.7 ± 15.8
0.29
a
O-C7A (°)
32.1 ± 11
39 ± 8.2
36.8 ± 7.6
36.3 ± 7.5
C2-7A (°)
10 ± 11.3
16.5 ± .9.7a
18.8 ± 12.2
17.8 ± 11.1
C2-5A (°)
10.3 ± 9.3
5.9 ± 9a
12.7 ± 7.4
C5-7A (°) O-C2A (°)
-0.3 ± 6.1 22.8 ± 9.1
10.5 ± 5a 22.5 ± .8.2
O-C1A (°)
-6.1 ± 7.3
C1-2 A (°)
29 ± 4.4
0.58
A vs B (post-OP) P value
0.19
\0.01
0.64
11.1 ± 5.9
0.26
\0.05
6.1 ± 7.4 17.9 ± 8.9
6.7 ± 7.7 18.6 ± 7.9
\0.001 \0.05
\0.01 0.06
-7 ± 6.9
-10.2 ± 7.6
-10.4 ± 6.9
\0.05
0.06
29.6 ± 4.8
28.1 ± 5.5
29 ± 5.8
0.50
0.55 0.05
St-SVA (mm)
25.3 ± 15
21.6 ± 12.7
16 ± 17.5
14.4 ± 15.8
\0.05
C7-s (°)
-19.4 ± 9.9
-25 ± 7.2a
-22.4 ± 6.5
-22.5 ± 6.5
0.17
0.30
SCA (°)
100.5 ± 8.7
106.9 ± 6.7a
105.8 ± 8.5
106.8 ± 7.1
\0.05
0.64
O-s (°)
13.6 ± 5.5
13.7 ± 5.2
14.5 ± 5.4
14.3 ± 5.4
0.51
0.64
C1-s (°)
19.6 ± 7.7
21.2 ± 7.7
25.1 ± 8
25.2 ± 7.7
\0.05
\0.05
C2-s (°)
-9.2 ± 8.9
-8.5 ± 9
-3.4 ± 9.5
-4.3 ± 7.7
\0.05
\0.05
C5-s (°)
-19.9 ± 8.9
-14.5 ± 8.2a
-16.2 ± 6.6
-15.4 ± 5.5
\0.05
0.35
a
Statistical significant change after surgery, pre-op pre-operative, post-op post-operative, A (group A) the value of C5-7A increased by more than 6°, B (group B) the value of C5-7A increased by less than 6° or decreased
operation, the C5-s showed upward inclination (5.4°), and C7-s showed downward inclination (5.6°). The summation of change values of the C5 and C7 slope (11°) was nearly the same as the regional change at the C5-7 level (10.8°). Patients in group B (those who maintained similar regional curvature at the C5-7 level) did not show any change in parameters post-operatively. Groups A and B had similar values for the upper cervical spine parameters (O-C2A, O-C1A) peri-operatively, when the O-slope was fixed.
Subgroups based on the St-SVA change (Table 4) There was no significant difference in age, gender, or duration of follow-up among the three groups perioperatively. Group D showed more kyphotic curvature segmentally (C5-7A, C2-5A) and globally (C2-7A) at the lower cervical spine and more lordotic curvature at the upper cervical spine (O-C2A; 25.8° vs 18.7° vs 12.9°) than groups S and I pre-operatively. The St-CVA value in group
123
123
-12.3 ± 7.9
-20.1 ± 7.4
C2-s (°)
C5-s (°)
14.5 ± 4.9 24.5 ± 7.8 -4.2 ± 8.7 -16.9 ± 7.1
-7.8 ± 8.8a -14.8 ± 7.2a
104.8 ± 6.4
-20.9 ± 5.4
13.9 ± 6.1
a
28.7 ± 4.8 16.1 ± 16.1
22.4 ± 7.1a
-23 ± 8.4
30.3 ± 4.9 18.3 ± 10.9a
a
14.4 ± 4.4
-15 ± 7.1
-5.3 ± 8.2
24.9 ± 7.3
-16.5 ± 7.6
1.2 ± 7.6
27.3 ± 7.4
14.1 ± 5.3
108 ± 9.2
-23.7 ± 8.2
-24 ± 6.1a 108.1 ± 5.4a
25.3 ± 5.8 12.2 ± 17.8
-12.4 ± 4.6
12.9 ± 6.3
7.2 ± 8.4
17.7 ± 7.2
24.9 ± 11.1
37.5 ± 10.4
11.7 ± 9.3
(7/4)
55.5 ± 8.3
I (Pre-OP) (n = 11)
29.7 ± 5.1 15.5 ± 15.3
-10 ± 6.5
19.7 ± 7.1
9.2 ± 6.1
9.7 ± 5.9
18.9 ± 8.2
38.3 ± 6.4a
S (post-OP)
-15.1 ± 6.9
-6.1 ± 9.6a
20.5 ± 10.4a
13.4 ± 5.7
106.5 ± 8.7
-24.8 ± 5.4
26.2 ± 5.8 23.6 ± 19.5a
-6.7 ± 6.3a
19.5 ± 7.6a
0.23
\0.01
\0.01
0.83
0.051
0.40
0.68 \0.05
\0.05
\0.05
0.17
\0.05
9 ± 6.6a 9.3 ± 6.9
\0.05
0.31
0.41
0.52
0.60
D vs S (pre-OP) P value
18.3 ± 11.3a
37.7 ± 8.3
I (Post-OP)
\0.001
\0.01
\0.05
0.20
0.58
0.57
0.65 0.14 0.77
\0.01
0.15
0.40
0.19
0.10
S vs I (pre-OP) P value
\0.05 \0.05
\0.001
\0.05
\0.01
\0.001
D vs W (pre-OP) P value
0.99
0.59
0.25
0.86
0.47
0.75
0.09 0.31
0.37
0.49
0.68
0.46
0.33
0.81
D vs S vs I (post -OP) P value
Statistically significant change after surgery, pre-op pre-operative, post-op post-operative, D (decreased group) the St-SVA value decreased by more than 5 mm, S (similar group) the St-SVA value decreased or increased by less than 5 mm, I (increased group) the St-SVA value increased by more than 5 mm
a
13.5 ± 6.1
17.8 ± 7.1
99 ± 9.7
SCA (°)
O-s (°)
-19.5 ± 11.6
C7-s (°)
C1-s (°)
105.8 ± 7.3
29.9 ± 4.3 29.7 ± 13.4
C1-2 A (°) St-SVA (mm)
-10 ± 6.4
-8.1 ± 7.9a
-4.1 ± 8.4
3.7 ± 6.3
12.7 ± 7.1
16.4 ± 10.3
35 ± 6.9
18.7 ± 7.7
a
11.3 ± 10
(11/15)
56 ± 8.7
S (pre-OP) (n = 26)
22.2 ± 9.6a
7.7 ± 7.3
a
6.9 ± 10.3
O-C1A (°)
C2-5A (°)
14.7 ± 11.9
-0.1 ± 7.2
7.3 ± 8.4
C2-7A (°)
36.9 ± 9.4a
25.8 ± 9.1
7.3 ± 11.2
O-C7A (°)
O-C2A (°)
32.3 ± 11.6
F/U (months)
D (post-OP)
C5-7A (°)
(13/11)
16.3 ± 18.1
Sex (M/F)
53.7 ± 7.8
Age (years)
D (pre-OP) (n = 24)
Table 4 Sequential change in cervical alignment according to St-SVA value outcomes (I: increased, S: similar, D: decreased)
Eur Spine J
Eur Spine J
D was also significantly larger than that of groups S and I pre-operatively. The curvature at the surgical level (C5-7A) was not significantly different (-0.1° vs 3.7°) between groups D and S; however, the upper adjacent level (C2-5A) was significantly lower in group D (7.3°) than group S (12.7°) pre-operatively. When comparing groups D and I, the curvature of both the surgical (C5-7) and upper adjacent level (C2-5) was also significantly different (P \ 0.05) pre-operatively. After ACDF, the C1 and C2 slopes of group D showed an upward inclination (4.5° and 4.6°), which was similar to the amount of C5-s change (5.3°). However, the inclination of the C1-s and C2-S of other groups (group S and I) was maintained or showed post-operative downward movement. After ACDF, none of the parameters in the three groups were significantly different regardless of the preoperative alignment of the cervical spine.
Discussion Horizontal gaze has a significant impact on activities of daily living and quality of life, and it can reflect the position of the skull [10]. In the present study, we adopted the McGregor line (O-s) as a substitute for the Frankfort horizontal (FH) line (the imaginary line between the lower border of the orbit and the external auditory canal) [11]. Although the Frankfort horizontal line is the representative cephalometric reference line to determine the degree of vertical gaze, we found that the anatomical margin of the external auditory canal (EAC) could be more ambiguous than the sellar turcica. Moreover, the McGregor line is also a good postural variable reflecting the position of the skull, and it can indicate the degree of vertical alteration in a person’s gaze. We also found a close relationship between the FH line and McGregor line in this study. If we fixed the O-s value peri-operatively (30 patients, randomly selected), the pre-operative value of the FH line slope remained similar after surgery (mean value, before surgery: 20.3°; after surgery: 20.4°). In general, sequential change in cranio-cervical alignment after ACDF appears to follow the pattern described below when the horizontal gaze is fixed (Table 1; Fig. 2). Insertion of the interbody graft results in vertical distractive forces to the surgical segment (C5-6-7), which induces upward movement of C5-s and downward movement of C7-s (D 2.9°). An upward movement of C5-s (D 3.2°) induces a decrease in C2-5A (D 3°) because C2-s does not change. A downward movement of C7-s determines the C7 reference point to a forward direction, which can induce an increase in the SCA (D 3.8°). C7-s was mathematically related to SCA (r = -0.75), and Le Huec has previously
described the formula: SCA = 90° - C7 s ? sellar turcica tilt [1]. Peri-operatively, none of the parameters related to the upper cervical region (O-C2A, C1-2A, C1-s and C2-s) changed when the horizontal gaze was fixed. This result may provide insight on the likelihood of upper cervical movement. First, maintaining horizontal gaze was not controlled by the O-C1 joint separately; rather, its function expanded to the C1-2 joint. Therefore, the O-C1-C2 segments are closely connected and act as a unified segment. Second, the degree of horizontal gaze can reflect upper cervical spine alignment (O-C2). This phenomenon is affirmed by the correlation between sagittal spinal parameters. Since the O-s was closely related to the C2-s (r = 0.51) and C1-S (r = 0.33), the difference in horizontal gaze can indicate the difference in the values of C1-s and C2-s. An upward horizontal gaze induced the upward movement of C1-s and C2-S concurrently, which may have induced backward movement of the center of gravity (sellar turcica), resulting in a decrease in the St- SVA (r = -0.36) value. However, parameters related to the surgical level (C5-7A, C7-s) were not significantly correlated with the change in the O-s value (r = -0.08, r = 0, respectively). After ACDF, the change in horizontal gaze was closely correlated with the upper adjacent segment (C2-5) and upper cervical segment (O-C2) rather than the surgical level (C5-7). To assess radiological outcomes after ACDF, we determined the degree of angle correction (D C5-7A) at the surgical level. Because the values of O-s and C7-s were similar before surgery, the pre-operative O-C7A value did not differ between the two groups (A and B) (P = 0.58). When analyzing the C2-7A as C2-5A and C5-7A, only C57A differed significantly between the two groups before surgery (P \ 0.001). After ACDF, the C2-7A of group A was similar to that of group B due to the improvement in C5-7A lordosis (D 10.8°). Sequentially, the C2-5A value of group A decreased (D 4.4°) in proportion to the upward inclination of C5-s (D 5.4°) and C2 (D 0.7°). Therefore, the ACDF procedure induced sequential change in cervical alignment mainly at the lower cervical spine (C2-C7) in group A, which showed segmental kyphosis (C5-7A) preoperatively. Patients in group B who had sufficient segmental lordosis at the surgical level (C5-7) and the upper adjacent level (C2-5) before surgery showed no significant differences perioperatively (P [ 0.05). Nonsignificant upward inclination of the C1-S (D 1.6°) and C2-s (D 0.7°) in group A and downward inclination of the C2-S (D 0.9°) in group B were found. We assumed that the St-SVA was another representative factor for estimating radiological outcome because it is well known that the degree of St-SVA is closely related to
123
Eur Spine J
Fig. 2 General sequential change in cranio-cervical alignment after ACDF, when the horizontal gaze is fixed. a Occipital-slope, b C5-slope, c C7-slope, d spino-cranial angle (SCA), e center of the sellar turcica-C7 sagittal vertical axis (St-SVA), - decrement, ? increment
the clinical outcome in terms of neck pain and healthrelated quality of life (HRQOL) [12, 13]. In general, preoperative spinal curvature differed among the three groups (D, S, and I), with the exception of O-C7A and C7-S. Group D had more kyphotic curvature than groups S and I at the lower cervical spine pre-operatively, and the St-SVA value was also greater in group D than in the other two groups. After ACDF, group D showed restored regional lordosis at the lower cervical spine (C5-7A: D 7.8°, C2-5A: D 1.6°) and reduced lordosis at the upper cervical spine (OC2A: 25.8° -[ 22.2°). The upward inclination of C1-s and C2-s induced a decrease in O-C2A. Therefore, the ACDF procedure in group D induced sequential change in cervical curvature that expanded to the upper cervical spine (O-C2), resulting in a significant decrease in the St-SVA value (Fig. 3). On the other hand, group S and I also showed restored segmental lordosis at the surgical level (C5-7A: D 5.5° and D 2.1°); however, the inclination of C1-s and C2-S was maintained similarly (group S) or showed a downward movement (group I) after surgery. Therefore, St-CVA was either unchanged (group S) or aggravated (group I). These results show that the pre-operative state of C2-5A (upper adjacent level), C1-s, and C2-s may be the key determinant of improvement in St-SVA. In other words, the
123
state of degeneration in the cervical spine may determine the radiological outcome after ACDF. If one patient (group D, relatively high degenerative state) shows regional kyphosis at the lower cervical spine (C2-5 and C5-7) and higher lordosis at the upper cervical spine (O-C2) before surgery, ACDF can restore regional lordosis at the lower cervical spine (C2-7A), which then induces kyphosis at the upper cervical spine. If another patient (group I, relatively non-degenerative) shows the opposite state of cervical curvature, ACDF can decrease regional lordosis at the lower cervical spine (C2-7A), which then induces lordotic change at the upper cervical spine. Ironically, although segmental lordosis at the surgical level (C5-7) was improved in group I, regional lordosis at C2-7A decreased after ACDF. Such a finding results from the downward movement of C2-s. We did not determine the exact physiologic origins of cervical spine changes after ACDF, and thus there may be additional undisclosed factors. Previous reports showed that preservation or secure more lordotic curvature at surgical level (Villavicencio et al.) and decreases in the SVA-value (Tang et al.) are the most important factors for achieving good clinical outcomes related to neck pain or HRQOL after cervical spine
Eur Spine J
Fig. 3 ACDF procedure in group D (St-SVA, decreased) induced sequential change in cervical curvature that expanded to the upper cervical spine. a Occipital-slope, b C5-slope, c C7-slope, d spino-
cranial angle (SCA), e center of the sellar turcica-C7 sagittal vertical axis (St-SVA), - decrement, ? increment
surgery [12, 13]. Therefore, the present study contributes to the theoretical basis of our understanding of neck pain and HRQOL after ACDF considering upward sequential change of the cervical spine.
alignment changes was not discussed nor was the influence of minor residual stenosis at C4-5 in a patient who underwent fusion at C5-7.
Conclusion Limitation and interpretation This study was retrospective, so it was not possible to control the cervical spine position in detail. Second, the investigation of sequential change of the cervical spine was confined to the specific surgical level (C5-6-7). Future studies of anterior cervical surgery with another lesion (C45-6), multi-levels (more than three level) and more extensive corrective surgery (corpectomy) would be informative [10]. Moreover, comparative study between plate-cage system and other surgical constructs (zero-profile with selflocking system or TDR) may be also interesting [8, 9, 14]. Third, the present study did not include the thoraco-lumbar region for assessment of the effect of trunk alignment on cervical spine alignment, although consecutive patients did not show sagittal imbalance pre-operatively in this study. Fourth, the impact of pain or residual disability on
Since the regional or global cervical angle is a relative value made by two different slope lines, determining the degree of the slope angle is the most crucial step for assessing a change in spinal alignment. ACDF can induce improvement in regional lordosis at the surgical level and can subsequently influence changes in the upper cervical segment through upward inclination of C1-s and C2-s according to pre-operative kyphotic change of the lower cervical spine. Then, sequential change in cervical spine alignment can influence the StCVA and SCA values. Compliance with ethical standards Conflicts of interest All authors certify that that they have no affiliations with or involvement in any organization or entity with any financial or non-financial interest in the subject of this paper.
123
Eur Spine J
References 1. Le Huec JC, Demezon H, Aunoble S (2015) Sagittal parameters of global cervical balance using EOS imaging: normative values from a prospective cohort of asymptomatic volunteers. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc 24:63–71. doi:10.1007/s00586-014-3632-0 2. Nojiri K, Matsumoto M, Chiba K, Maruiwa H, Nakamura M, Nishizawa T, Toyama Y (2003) Relationship between alignment of upper and lower cervical spine in asymptomatic individuals. J Neurosurg 99:80–83 3. Lee SH, Kim KT, Seo EM, Suk KS, Kwack YH, Son ES (2012) The influence of thoracic inlet alignment on the craniocervical sagittal balance in asymptomatic adults. J Spinal Disord Tech 25:E41–E47. doi:10.1097/BSD.0b013e3182396301 4. Toyama Y, Matsumoto M, Chiba K, Asazuma T, Suzuki N, Fujimura Y, Hirabayashi K (1994) Realignment of postoperative cervical kyphosis in children by vertebral remodeling. Spine 19:2565–2570 5. Matsunaga S, Onishi T, Sakou T (2001) Significance of occipitoaxial angle in subaxial lesion after occipitocervical fusion. Spine 26:161–165 6. Lee CH, Kim KJ, Hyun SJ, Yeom JS, Jahng TA, Kim HJ (2015) Subsidence as of 12 months after single-level anterior cervical inter-body fusion. Is it related to clinical outcomes? Acta Neurochirurgica. doi:10.1007/s00701-015-2388-6 7. Lee YS, Kim YB, Park SW (2014) Risk factors for postoperative subsidence of single-level anterior cervical discectomy and fusion: the significance of the preoperative cervical alignment. Spine 39:1280–1287. doi:10.1097/brs.0000000000000400 8. Grasso G, Giambartino F, Tomasello G, Iacopino G (2014) Anterior cervical discectomy and fusion with ROI-C peek cage: cervical alignment and patient outcomes. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc 23(Suppl 6):650–657. doi:10.1007/s00586-014-3553-y
123
9. Nemoto O, Kitada A, Naitou S, Tachibana A, Ito Y, Fujikawa A (2015) Stand-alone anchored cage versus cage with plating for single-level anterior cervical discectomy and fusion: a prospective, randomized, controlled study with a 2-year follow-up. Eur J Orthop Surg Traumatol Orthopedie Traumatologie 25(Suppl 1):S127–S134. doi:10.1007/s00590-014-1547-4 10. Mattei TA, Rehman AA, Issawi A, Fassett DR (2015) Surgical challenges in the management of cervical kyphotic deformity in patients with severe osteoporosis: an illustrative case of a patient with Hajdu-Cheney syndrome. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. doi:10. 1007/s00586-015-4092-x 11. Lundstrom A, Lundstrom F (1995) The Frankfort horizontal as a basis for cephalometric analysis. Am J Orthodont Dentofac Orthop Off Publ Am Assoc Orthodont Const Soc Am Board Orthodont 107:537–540 12. Tang JA, Scheer JK, Smith JS, Deviren V, Bess S, Hart RA, Lafage V, Shaffrey CI, Schwab F, Ames CP (2015) The impact of standing regional cervical sagittal alignment on outcomes in posterior cervical fusion surgery. Neurosurgery 76(Suppl 1):S14– S21. doi:10.1227/01.neu.0000462074.66077.2b (discussion S21) 13. Villavicencio AT, Babuska JM, Ashton A, Busch E, Roeca C, Nelson EL, Mason A, Burneikiene S (2011) Prospective, randomized, double-blind clinical study evaluating the correlation of clinical outcomes and cervical sagittal alignment. Neurosurgery 68:1309–1316. doi:10.1227/NEU.0b013e31820b51f3 (discussion 1316) 14. Gao Y, Liu M, Li T, Huang F, Tang T, Xiang Z (2013) A metaanalysis comparing the results of cervical disc arthroplasty with anterior cervical discectomy and fusion (ACDF) for the treatment of symptomatic cervical disc disease. J Bone Jt Surg Am 95:555–561. doi:10.2106/jbjs.k.00599
