Eur Spine J DOI 10.1007/s00586-015-3911-4

ORIGINAL ARTICLE

Comparison of anterior approach versus posterior approach for the treatment of multilevel cervical spondylotic myelopathy Jiaquan Luo1 • Kai Cao2 • Sheng Huang1 • Liangping Li1 • Ting Yu1 • Cong Cao1 • Rui Zhong1 • Ming Gong1 • Zhiyu Zhou1 • Xuenong Zou1

Received: 26 December 2014 / Revised: 23 March 2015 / Accepted: 26 March 2015 Ó Springer-Verlag Berlin Heidelberg 2015

Abstract Purpose The purpose of this study is to evaluate the clinical outcomes, complications, and surgical trauma between anterior and posterior approaches for the treatment of multilevel cervical spondylotic myelopathy (CSM). Study design Systematic review and meta-analysis. Methods Randomized controlled trials or non-randomized controlled trials published up to November 2014 that compared the clinical effectiveness of anterior and posterior surgical approaches for the treatment of multilevel CSM were acquired by a comprehensive search in four electronic databases (PubMed, EMBASE, Cochrane Controlled Trial Register and MEDLINE). Exclusion criteria were non-controlled studies, combined anterior and posterior surgery and cervical myelopathy caused by ossification of the posterior longitudinal ligament. The main end points included: recovery rate; Japanese Orthopedic Association (JOA) score; complication rate; reoperation rate; blood loss; operation time and length of stay. Results A total of ten studies were included in the metaanalysis; none of which were randomized controlled trials. All of the selected studies were of high quality as indicated J. Luo, K. Cao and S. Huang equally contributed to this work. & Xuenong Zou [email protected] Jiaquan Luo [email protected] 1

Department of Spine Surgery/Orthopaedic Research Institute, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou 510080, People’s Republic of China

2

Department of Orthopaedics, The First Affiliated Hospital of Nanchang University, Nanchang 330006, People’s Republic of China

by the Newcastle–Ottawa scale. In six studies involving 467 patients, there was no significant difference in the preoperative JOA score between the anterior surgery group and the posterior group [P [ 0.05, WMD -0.00 (-0.50, 0.50)]. In four studies involving 268 patients, the postoperative JOA score was significantly higher in the anterior surgery group compared with the posterior surgery group [P \ 0.05, WMD 0.79 (0.16, 1.42)]. In five studies involving 420 patients, there was no statistically significant difference in recovery rate between the anterior and posterior surgery groups [P [ 0.05, WMD 2.73 (-8.69, 14.15)]. In nine studies involving 804 patients, the postoperative complication rate was significantly higher in the anterior surgery group compared with the posterior surgery group [P = 0.009, OR 1.65 (1.13, 2.39)]. In five studies involving 294 patients, the reoperation rate was significantly higher in the anterior surgery group compared with the posterior surgery group [P = 0.0001, OR 8.67 (2.85, 26.34)]. In the four studies involving 252 patients, the intraoperative blood loss and operation time was significantly higher in the anterior surgery group compared with the posterior surgery group [P \ 0.05, WMD -40.25 (-76.96, -3.53) and P \ 0.00001, WMD 61.3 (52.33, 70.28)]. In the three studies involving 192 patients, the length of stay was significantly lower in the anterior surgery group compared with the posterior surgery group [P \ 0.00001, WMD -1.07 (-2.23, -1.17)]. Conclusions In summary, our meta-analysis suggested that a definitive conclusion could not be reached regarding which surgical approach is more effective for the treatment of multilevel CSM. Although anterior approach was associated with better postoperative neural function than posterior approach in the treatment of multilevel CSM, there was no apparent difference in the neural function recovery rate between the two approaches. Higher rates of surgery-

123

Eur Spine J

related complication and reoperation should be taken into consideration when anterior approach is used for patients with multilevel CSM. Keywords Anterior approach  Posterior approach  CSM  Meta-analysis

At present, no standards or guidelines exist for the treatment of multilevel CSM; we performed the current meta-analysis to compare the effectiveness of anterior and posterior procedure for patients with multilevel CSM in terms of neurological function outcome, surgical complications, reoperation rate, and surgical trauma.

Introduction

Materials and methods

Cervical spondylotic myelopathy (CSM) is the most common cause of spinal cord dysfunction [1]. This is usually a direct consequence of circumferential compression because of degenerative processes, often in a congenitally narrowed spinal canal. The degeneration of the intervertebral disc, uncovertebral joint, facet joint, posterior longitudinal ligament, and ligamentum flavum causes spinal cord compression and cervical myelopathy [2]. To great extent, these injuries and compression lead to direct alteration or injury of neurologic pathways and cause pathophysiologic changes similar to traumatic spinal injury [3]. Surgical approaches for CSM can be broadly divided into the anterior, posterior and combined anterior and posterior. Anterior approaches usually involve anterior cervical discectomy with fusion (ACDF) or anterior cervical corpectomy with fusion (ACCF), whereas laminectomy with or without instrumentation and laminoplasty represent the posterior approach. However, there is considerable controversy over which surgical approach will lead to the best clinical outcome in multilevel CSM [4, 5]. With respect to anterior approach, ACDF and ACCF can establish a solid cervical stability that is conducive to relieving pressure on the level of compressed spinal cord. However, the reconstruction of the cervical spine after multisegment ACDF or ACCF is technically demanding for the surgeon and more bone grafts are needed for fusion, resulting in an increased rate of graft-, instrumentation-, and surgery-related complications. In addition, some authors have argued that anterior surgery also appears to be associated with a high rate of adjacent degeneration [6, 7]. As for posterior approach, decompression via posterior laminectomy or laminoplasty and cervical lordosis alignment allow the spinal cord to float away from ventral compression, thus receiving an indirect decompression of the whole cervical cord. However, if posterior shift of the cord is not sufficient, ventral constriction of the cord may persist, leading to decreased recovery from myelopathy. Some authors have recognized the complications of this procedure, including axial pain and C5 root palsy [8–10]. Further, some studies have reported that patients who undergo posterior laminoplasty or laminectomy are more likely to suffer from cervical kyphosis or instability than those who receive treatment via anterior approaches [11–13].

Search strategy and study selection

123

We searched for studies published between January 1960 and November 2014 that compared clinical effectiveness of anterior and posterior surgical approaches for the treatment of multilevel CSM. The databases included PubMed, EMBASE, Cochrane Controlled Trial Register and MEDLINE with no language restriction. The following search terms were used: (1) cervical myelopathy OR CSM OR myelopathy OR cervical spondylosis OR cervical vertebrae OR cervical stenosis; (2) corpectomy OR ACDF OR anterior cervical discectomy and fusion OR anterior decompression and fusion OR anterior decompression OR ventral decompression OR ventral approach OR ventral; (3) laminoplasty OR laminectomy OR posterior decompression OR posterior decompression and fusion OR dorsal decompression OR dorsal approach OR dorsal; (1) and (2) or (3). Reference lists of all included studies were scanned to identify additional potentially relevant studies. Two reviewers independently screened the titles and abstracts of identified papers, and full text copies of all potentially relevant studies were obtained. Inclusion criteria Studies were included if they met the following criteria: (1) study design: randomized or non-randomized controlled study; (2) study population: patients with CSM caused by multisegmental spinal stenosis (C2 segments), excluding tumors, trauma, soft disc herniation or previous surgery; (3) purpose of interventions: to compare clinical outcome differences between posterior cervical canal decompression and anterior cervical canal decompression (regardless of the specific surgical approach); (4) outcome measurements: neurological recovery outcomes, surgical complications, reoperation rate, operation time and blood loss. Studies that did not meet the above criteria were excluded from selection. Data extraction The following information was extracted from each study: (1) study ID; (2) study design; (3) study location; (4) main inclusion/exclusion criteria; (5) patient demographics; (6)

Eur Spine J

length of follow-up; (7) number of surgical segments; (8) surgical approach for each group; (9) JOA scores before and after surgery; (10) recovery rate; (11) reoperation rate; (12) number of complications, type of complications, and rate of complications; and (13) operation time and blood loss. Data analysis We performed all meta-analyses with the Review Manager software (RevMan Version 5.1; The Nordic Cochrane Center, The Cochrane Collaboration, Copenhagen, Denmark). Heterogeneity was tested using Chi-square test and quantified by calculating I2 statistic, for which P \ 0.1 and I2 [ 50 % was considered to be statistically significant. For the pooled effects, weighted mean difference (WMD) or standard mean difference (SMD) was calculated for continuous variables according to the consistency of measurement units, and odds ratio (OR) was calculated for dichotomous variables. Continuous variables are presented as mean differences and 95 % confidence intervals (CI), whereas dichotomous variables are presented as odds ratios and 95 % CI. Random-effects or fixed-effects models were used depending on the heterogeneity of the studies included.

2546 records identified through electronic and manual searching

115 full-text articles were included in full text

2431 references were excluded by screening the titles,abstracts, duplicates, irrelevant studies,case reports, not comparative studies and review

106 full-text articles excluded with reasons 10 full-text articles assessed for eligibility

10 studies included in meta-analysis

Fig. 1 The flow chart shows the article selection process we performed

(n = 467 patients; 226 in the anterior surgery group and 241 in posterior surgery group). There was no significant difference in the preoperative JOA score between the anterior surgery group and the posterior group [P [ 0.05, WMD -0.00 (-0.50, 0.50); Fig. 2].

Results Postoperative JOA score The process of identifying relevant studies is summarized in Fig. 1. From the selected databases, 2546 references were obtained. By screening the titles and abstracts, 2431 references were excluded due to duplicates, irrelevant studies, case reports, not comparative studies and review. The remaining 115 reports underwent a detailed and comprehensive evaluation. Finally, ten studies were included in this meta-analysis [14–23]. Table 1 summarizes the baseline characteristics assessment. Quality assessment of the studies The major baseline characteristics of participants in each study were similar. Newcastle–Ottawa quality assessment scale was used to assess the quality of included studies [24]. Of the studies, eight scored 8 points and one scored 7 points; hence, the studies were of a relatively high quality (Table 2). Clinical outcome Preoperative JOA score Six studies used the JOA score to assess the clinical outcome, all of which provided a preoperative JOA score

Four of six studies provided a postoperative (last followup) JOA score (n = 268 patients; 128 in the anterior surgery group and 140 in the posterior surgery group). The postoperative JOA score was significantly higher in the anterior surgery group compared with the posterior surgery group [P \ 0.05, WMD 0.79 (0.16, 1.42); Fig. 3]. Recovery rate Recovery rate obtained in five studies with a total of 420 patients (203 in the anterior surgery group and 217 in the posterior surgery group) was analyzed. There was no statistically significant difference in recovery rate between the anterior and posterior surgery groups [P [ 0.05, WMD 2.73 (-8.69, 14.15); Fig. 4]. Complications Complications obtained in nine studies with a total of 809 patients (435 in the anterior surgery group and 374 in the posterior surgery group) were analyzed. The postoperative complication rate was significantly higher in the anterior surgery group compared with the posterior surgery group [P = 0.009, OR 1.65 (1.13, 2.39); Fig. 5].

123

Eur Spine J Table 1 Characteristics of included studies References Liu et al. [14]

Design

Country

Non-randomized

China

Prospective

Number of patient

Patient age

Surgical approach

Total: 52

A 54.64 ± 11.49

A ACDF

A 25, P 27

P 57.33 ± 10.09

P laminoplasty

Total: 50

A 60

A ACDF

A 28, P 22

P 64

P laminectomy

Controlled Ghogawala et al. [15]

Non-randomized

USA

Prospective Pilot Kristof et al. [16]

Non-randomized

Germany

Retrospective

Total: 103

A 62.5 ± 10.61

A ACCF

A 42, P 61

P 66.0 ± 12.4

P laminectomy

Total: 26

A 53

A ACCF

A 13, P 13

P 54

P laminoplasty

Total: 83

A 60.4 ± 8.4

A ACCF

A 34, P 49

P 64.8 ± 11.7

P laminoplasty

Total: 83

A 56.0 ± 11.5

A ACCF

A 41, P 42

P 54.3 ± 8.9

P laminoplasty

Total: 86

A 59.2 ± 10.7

A ACCF

A 39, P 47

P 61.2 ± 10.1

P laminoplasty

Total: 47

A 52.7 ± 7.6

A ACCF

A 23, P 24

P 56.5 ± 11.2

P laminoplasty

Comparative Charles et al. [17]

Non-randomized

USA

Retrospective Matched-cohort Shibuya et al. [18]

Non-randomized

Japan

Retrospective Yonenobu et al. [19]

Non-randomized

Japan

Retrospective Comparative Hirai et al. [20]

Non-randomized

Japan

Prospective Comparative Wada et al. [21]

Non-randomized

Japan

Retrospective Comparative Seng et al. [22]

Non-randomized

Singapore

Prospective

Total: 116

A 58.6 ± 10.7

A ACCF OR ACDF

A 64, P 52

P 60.6 ± 10.8

P laminoplasty

Total: 264 A 169, P 95

A 52.53 ± 10.6 P 62.83 ± 10.68

A ACCF or ACDF P laminoplasty or laminectomy

Comparative Fehlings et al. [23]

Prospective Observational

USA

ACDF anterior cervical discectomy with fusion, ACCF anterior cervical corpectomy with fusion, A anterior approaches, P posterior approaches

Table 2 Quality assessment of included studies in the metaanalysis according to the Newcastle–Ottawa scale

References

Selection

Comparability

Exposure

Total score 8

Liu et al. [14]

3

2

3

Ghogawala et al. [15]

3

2

3

8

Kristof et al. [16]

3

2

3

8

Charles et al. [17]

3

2

3

8

Shibuya et al. [18]

2

2

3

7

Yonenobu et al. [19]

3

2

3

8

Hirai et al. [20]

3

2

3

8

Wada et al. [21]

3

2

3

8

Seng et al. [22]

3

2

3

8

Fehlings et al. [23]

3

2

3

8

Reoperation rate Reoperation rate obtained in five studies with a total of 294 patients (134 in the anterior surgery group and 160 in the

123

posterior surgery group) was analyzed. The reoperation rate was significantly higher in the anterior surgery group compared with the posterior surgery group [P = 0.0001, OR 8.67 (2.85, 26.34); Fig. 6].

Eur Spine J Anterior Study or Subgroup Hirai et al.[20] Liu T et al.[14]

Mean

Posterior

SD Total Mean 3.1

39

8.16 3.41

25

9.9

Mean Difference

SD Total Weight 2.9

47

8.59 2.98

27

9.7

15.2%

IV, Fixed, 95% CI

0.20 [-1.08, 1.48]

8.1% -0.43 [-2.18, 1.32]

Seng C et al.[22]

11

3

64

11

3

52

20.6%

0.00 [-1.10, 1.10]

Shibuya S et al.[18]

8.6

2.9

34

7.9

2.4

49

17.7%

0.70 [-0.48, 1.88]

Wada E et al.[21]

7.9

1.8

23

7.4

2.2

24

18.9%

0.50 [-0.65, 1.65]

Yonenobu K et al.[19]

8.2

2.2

41

9.3

3

42

19.5% -1.10 [-2.23, 0.03]

241 100.0% -0.00 [-0.50, 0.50]

226

Total (95% CI)

Mean Difference

IV, Fixed, 95% CI

Heterogeneity: Chi² = 6.04, df = 5 (P = 0.30); I² = 17%

-100

Test for overall effect: Z = 0.00 (P = 1.00)

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 2 Forest plot of preoperative JOA score between the anterior surgery group and the posterior group. CI confidence interval, M-H Mantel– Haenszel, SD standard deviation

Anterior Study or Subgroup Hirai et al.[20]

Mean 15

Posterior

Mean Difference

SD Total Mean SD Total Weight 47

36.9%

27

18.2% -0.47 [-1.94, 1.00]

3

24

14.4%

1.20 [-0.46, 2.86]

12.8 2.7

42

30.5%

0.50 [-0.64, 1.64]

140 100.0%

0.79 [0.16, 1.42]

13.2 2.72

Wada E et al.[21]

13.4

2.8

23

12.2

Yonenobu K et al.[19]

13.3

2.6

41 128

Total (95% CI)

IV, Fixed, 95% CI

1.50 [0.46, 2.54]

13.5 2.6

25 13.67 2.7

39

2.3

Liu T et al.[14]

Mean Difference

IV, Fixed, 95% CI

Heterogeneity: Chi² = 5.09, df = 3 (P = 0.17); I² = 41%

-100

Test for overall effect: Z = 2.47 (P = 0.01)

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 3 Forest plot of postoperative JOA score between the anterior surgery group and the posterior group at 24 months postoperatively. CI confidence interval, M-H Mantel–Haenszel, SD standard deviation

Anterior Study or Subgroup Hirai et al.[20] Liu T et al.[14]

Mean 72.9

Posterior

SD Total Mean 28.3

59.79 23.43

Mean Difference

SD Total Weight

22.70 [11.01, 34.39]

26.6

47

20.9%

25 59.54 29.37

27

18.7%

0.25 [-14.14, 14.64] -1.20 [-14.97, 12.57]

39

50.2

Seng C et al.[22]

51.9

30.9

64

53.1

42.3

52

19.2%

Shibuya S et al.[18]

52.4

28.1

34

50.9

25.9

49

20.7%

1.50 [-10.41, 13.41]

Yonenobu K et al.[19]

44.9

26.2

41

55.3

30.2

42

20.5%

-10.40 [-22.55, 1.75]

217 100.0%

2.73 [-8.69, 14.15]

Total (95% CI)

203

Mean Difference IV, Random, 95% CI

IV, Random, 95% CI

Heterogeneity: Tau² = 127.11; Chi² = 16.13, df = 4 (P = 0.003); I² = 75%

-100

Test for overall effect: Z = 0.47 (P = 0.64)

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 4 Forest plot of recovery rate between the anterior surgery group and the posterior group at 24 months postoperatively. CI confidence interval, M-H Mantel–Haenszel, SD standard deviation

Blood loss

Operation time

Intraoperative blood loss obtained in four studies with a total of 252 patients (168 in the anterior surgery group and 184 in the posterior surgery group) was analyzed. The intraoperative blood loss was significantly higher in the anterior surgery group compared with the posterior surgery group [P \ 0.05, WMD -40.25 (-76.96, -3.53); Fig. 7].

Operation time obtained in four studies with a total of 252 patients (168 in the anterior surgery group and 184 in the posterior surgery group) was analyzed. The operation time was significantly higher in the anterior surgery group compared with the posterior surgery group [P \ 0.00001, WMD 61.3 (52.33, 70.28); Fig. 8].

123

Eur Spine J Anterior

Posterior

Odds Ratio

Events Total Events Total Weight

Study or Subgroup

0.7% 27.00 [2.56, 284.70]

9

13

1

13

Fehlings MG et al.[23]

20

160

17

86

44.0%

0.58 [0.29, 1.18]

Ghogawala Z et al.[15]

5

28

3

22

6.3%

1.38 [0.29, 6.52] 3.21 [0.77, 13.37]

Charles C et al.[17]

7

39

3

47

5.1%

17

42

22

61

24.3%

1.21 [0.54, 2.70]

Liu T et al.[14]

9

25

3

27

4.2%

4.50 [1.05, 19.22]

Seng C et al.[22]

3

64

2

52

4.8%

1.23 [0.20, 7.65]

Wada E et al.[21]

8

23

4

24

5.8%

2.67 [0.67, 10.54]

12

41

3

42

4.8%

5.38 [1.39, 20.82]

374 100.0%

1.65 [1.13, 2.39]

Hirai et al.[20] Kristof RA et al.[16]

Yonenobu K et al.[19]

435

Total (95% CI)

58

90

Total events

Odds Ratio M-H, Fixed, 95% CI

M-H, Fixed, 95% CI

Heterogeneity: Chi² = 20.58, df = 8 (P = 0.008); I² = 61%

0.01

Test for overall effect: Z = 2.62 (P = 0.009)

0.1

1

10

100

Favours [anterior] Favours [posterior]

Fig. 5 Forest plot of complications between the anterior surgery group and the posterior group

Anterior Study or Subgroup

Posterior

Odds Ratio

Events Total Events Total Weight

Charles C et al.[17]

0

13

1

13

49.4%

0.31 [0.01, 8.30]

Hirai et al.[20]

1

39

0

47

15.0%

3.70 [0.15, 93.45]

14.2%

8.56 [0.42, 174.46]

Liu T et al.[14] Shibuya S et al.[18] Wada E et al.[21]

3

25

0

27

10

34

0

49

9.9% 42.43 [2.39, 754.36]

7

23

0

24

11.5% 22.27 [1.19, 417.10]

1

21

Total events

8.67 [2.85, 26.34]

160 100.0%

134

Total (95% CI)

Odds Ratio M-H, Fixed, 95% CI

M-H, Fixed, 95% CI

Heterogeneity: Chi² = 5.78, df = 4 (P = 0.22); I² = 31%

0.01

Test for overall effect: Z = 3.81 (P = 0.0001)

0.1

1

10

100

Favours [Anterior] Favours [Posterior]

Fig. 6 Forest plot of reoperation rate between the anterior surgery group and the posterior group

Anterior Study or Subgroup Hirai et al.[20] Kristof RA et al.[16]

Mean 340

Posterior

SD Total

Mean

39

188

287

743.33 748.9

Mean Difference

SD Total Weight 92.1

47

42 838.83 876.6

61

15.3%

152.00 [58.16, 245.84]

101

54

64

183

141

52

82.0%

-82.00 [-122.54, -41.46]

Wada E et al.[21]

986

751

23

608

212

24

1.3%

378.00 [59.58, 696.42]

184 100.0%

-40.25 [-76.96, -3.53]

168

IV, Fixed, 95% CI

1.4% -95.50 [-411.24, 220.24]

Seng C et al.[22]

Total (95% CI)

Mean Difference

IV, Fixed, 95% CI

Heterogeneity: Chi² = 26.94, df = 3 (P < 0.00001); I² = 89%

-100

Test for overall effect: Z = 2.15 (P = 0.03)

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 7 Forest plot of blood loss between the anterior surgery group and the posterior group

Length of stay Length of stay obtained in three studies with a total of 192 patients (105 in the anterior surgery group and 87 in the

123

posterior surgery group) was analyzed. The length of stay was significantly lower in the anterior surgery group compared with the posterior surgery group [P \ 0.00001, WMD -1.07 (-2.23, -1.17); Fig. 9].

Eur Spine J Anterior Study or Subgroup Hirai et al.[20]

Mean

Posterior

SD Total

Mean

39

149

211 55.3

Mean Difference

SD Total Weight

Mean Difference

38.7

47

19.0%

62.00 [41.42, 82.58]

42 183.84 46.62

61

17.2%

45.40 [23.78, 67.02]

Seng C et al.[22]

186

38

64

123

28

52

55.7%

63.00 [50.98, 75.02]

Wada E et al.[21]

264

65

23

182

43

24

Kristof RA et al.[16]

229.24 60.1

8.0% 82.00 [50.35, 113.65]

184 100.0%

168

Total (95% CI)

IV, Fixed, 95% CI

IV, Fixed, 95% CI

61.30 [52.33, 70.28]

Heterogeneity: Chi² = 3.80, df = 3 (P = 0.28); I² = 21%

-100

Test for overall effect: Z = 13.39 (P < 0.00001)

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 8 Forest plot of operation time between the anterior surgery group and the posterior group

Anterior Study or Subgroup

Posterior

Mean Difference

Mean SD Total Mean SD Total Weight 5

0

13

3

0

13

Not estimable

Ghogawala Z et al.[15]

2.6

0

28

4

0

22

Not estimable

Seng C et al.[22]

3.7 1.5

64

Charles C et al.[17]

Total (95% CI)

105

5.4 1.4

Mean Difference IV, Fixed, 95% CI

IV, Fixed, 95% CI

52 100.0% -1.70 [-2.23, -1.17] 87 100.0% -1.70 [-2.23, -1.17]

Heterogeneity: Not applicable Test for overall effect: Z = 6.30 (P < 0.00001)

-100

-50

0

50

100

Favours [anterior] Favours [posterior]

Fig. 9 Forest plot of length of stay between the anterior surgery group and the posterior group

Discussion ACDF or ACCF and posterior laminectomy or laminoplasty have been reported for the treatment of multilevel CSM. There are several established options in the treatment of multilevel CSM in the form of anterior and posterior approaches. However, due to a lack of comprehensive studies comparing the clinical outcomes of both surgical approaches, the ideal surgical strategy for multilevel CSM remains controversial. Recently, Zhu et al. [25] reported that the anterior approach was associated with better postoperative neural function than the posterior approach in the treatment of multilevel CSM. There was no apparent difference in the neural function recovery rate. The complication and reoperation rates were significantly higher in the anterior group compared with the posterior group. However, the meta-analysis was based on eight studies. In addition, the relevant literature was only searched up to July 2012. Over the last 2 years, a series of comparative studies on surgical approaches for the treatment of multilevel CSM have been published; therefore, it remains necessary to verify above conclusion based on the latest high-quality studies for strong evidence. The objective of our meta-analysis is to systematically compare surgical approaches (anterior or posterior) for the treatment of multilevel CSM.

In our meta-analysis, from the literature search up to December 2014, eight articles were considered to be of sufficient methodological quality and were included. Metaanalysis results showed that no statistically significant difference was found in preoperative JOA score and recovery rate between the anterior and posterior surgery groups. Compared with the posterior surgery, the intraoperative blood loss and operation time, postoperative JOA score, postoperative complication rate were significantly higher, whereas the length of stay was significantly lower in the anterior surgery group. These are similar to the results reported by others [25]. A few meta-analyses have been published regarding decision making in the treatment of multilevel CSM. Liu et al. [26] published systematic review and meta-analysis of cohort studies comparing surgical treatment for CSM. In this article, the authors focused on comparing the clinical outcomes with different surgical approaches (ACDF and laminoplasty). Liu et al. [26] suggest that definitive conclusion could not be reached regarding which surgical approach is more effective for the treatment of multilevel CSM. Although ACDF was associated with better postoperative neural function than posterior laminoplasty in the treatment of multilevel CSM, there was no apparent difference in the neural function recovery rate between the two approaches [26]. Lawrence et al. [27] also

123

Eur Spine J

demonstrated that there is no clear advantage to either an anterior surgical approach or a posterior surgical approach when treating patients with multilevel CSM for both effectiveness and safety. Results of another study showed that anterior corpectomy and fusion are recommended for the treatment of multilevel CSM when the involved surgical segments were \3 [28]. Given the higher rates of surgical complications and complication-related reoperation and the higher surgical trauma associated with multilevel anterior corpectomy, however, it is suggested that posterior laminoplasty may be the preferred method of treatment for multilevel CSM when the involved surgical segments were equal to 3 or more [28]. The advantage for anterior surgery is that it is more radical than posterior surgery in decompressing the nerve tissue by directly removing all of the anterior pathogenic structures such as protruded discs, osteophyte or ossification lesion. With grafting, immediate stability of cervical spine can be achieved. In addition, with anterior stabilization, there is no longer concern over buckling of ligamentum flavum posteriorly. In our meta-analysis, it was shown that the anterior surgery was significantly better than posterior in the final follow-up postoperative JOA score. Anterior surgery seems to have better clinical outcomes. However, we also found that there was no statistically difference in the recovery rate between the anterior and posterior surgery groups. Previous study reported by Liu et al. [29] also found similar results. Liu et al. [29] conclude that anterior surgery had better clinical outcomes and more complications at the early stage after operation for both multilevel CSM and OPLL patients. Xu et al. [30] reported that definitive conclusions regarding the surgical treatment of multilevel ossification of posterior longitudinal ligament could not be made. Multilevel corpectomy and fusion are more technical and have a significantly higher rate of cerebrospinal fluid leakage and graft, instrumentation-related complications. We selected the complication rate for meta-analysis in the evaluation of complication-related outcomes. We found the postoperative complication rate was significantly higher in the anterior surgery group compared with the posterior surgery. For CSM patients, the main complications for anterior group were complications related to bone grafting, causing major concern during the postoperative course, dysphagia and dysphonia. This indicates that anterior approaches for the treatment of multilevel CSM are associated with a higher incidence of complications. These are consistent with the results reported by Seng et al. [22] also found that complications were higher for anterior surgery group. We also assessed reoperation rate as another complication-related measure. We found that the reoperation rate was significantly higher in the anterior surgery group compared with the posterior surgery group. Shibuya

123

et al. [18] reported ten out of 34 patients in the subtotal corpectomy group required reoperation because of pseudoarthrosis and adjacent level disc degeneration. Although the reasons for reoperation were different, anterior approach seemed to have a high risk of reoperation for the treatment of multilevel CSM. To evaluate the surgical trauma, length of stay, operation time and blood loss were selected for meta-analysis. Meta-analyses revealed that blood loss and operation time were significantly higher, whereas the length of stay was significantly lower in the anterior surgery group compared with the posterior surgery group. Similarly, Zhu et al. [25] also found that blood loss and operation time were significantly higher in the anterior subtotal corpectomy group compared with the posterior laminoplasty/laminectomy group. We believe that our result of meta-analysis is affected by several reasons. First, in this meta-analyses, all the studies selected were not RCT, while it did not influence the credibility of the results, because almost all the studies reported the baseline characteristics were matched for each group and the results for the meta-analysis of baseline characteristics also showed no significant difference between the two groups. Second, there was variability among the studies in the choice of indicators to evaluate the postoperative clinical effect. This clearly reflects the lack of a gold standard outcome measure. Third, there was a variable length of follow-up between the studies and this is particularly important for evaluating surgery results. Finally, clinical heterogeneity might be caused by the various indications for surgery and the surgical technologies used at the different treatment centers. Due to these limitations, the combined results of this meta-analysis should be cautiously accepted, and high-quality RCTs with long-term follow-up and large sample size are needed. In summary, our meta-analysis confirmed that the anterior approach was associated with better postoperative neural function than the posterior approach in the treatment of multilevel CSM. There was no significant difference in the neural function recovery rate. The complication, blood loss and operation time were significantly higher in the anterior approach compared with the posterior approach. The length of stay was significantly lower in the anterior approach compared with the posterior approach.

Conclusions In conclusion, our meta-analysis suggested that a definitive conclusion could not be reached regarding which surgical approach is more effective for the treatment of multilevel CSM. Although anterior approach was associated with better postoperative neural function than posterior approach in the

Eur Spine J

treatment of multilevel CSM, there was no apparent difference in the neural function recovery rate between the two approaches. Higher rates of surgery-related complication and reoperation should be taken into consideration when anterior approach is used for patients with multilevel CSM. Acknowledgments The present study was financially supported by National Program on Key Basic Research Project (973 Program; Grant No. 2012CB619105). Conflict of interest The authors declare that they have no conflicts of interest concerning this article.

References 1. Young WF (2000) Cervical spondylotic myelopathy: a common cause of spinal cord dysfunction in older persons. Am Fam Physician 62(1064–1070):1073 2. Edwards CN, Riew KD, Anderson PA, Hilibrand AS, Vaccaro AF (2003) Cervical myelopathy. current diagnostic and treatment strategies. Spine J 3:68–81 3. Baptiste DC, Fehlings MG (2006) Pathophysiology of cervical myelopathy. Spine J 6:190S–197S 4. Zhu B, Xu Y, Liu X, Liu Z, Dang G (2013) Anterior approach versus posterior approach for the treatment of multilevel cervical spondylotic myelopathy: a systemic review and meta-analysis. Eur Spine J 22:1583–1593. doi:10.1007/s00586-013-2817-2 5. Hirabayashi K, Bohlman HH (1995) Multilevel cervical spondylosis. Laminoplasty versus anterior decompression. Spine (Phila Pa 1976) 20:1732–1734 6. Bartolomei JC, Theodore N, Sonntag VK (2005) Adjacent level degeneration after anterior cervical fusion: a clinical review. Neurosurg Clin N Am 16:575–587. doi:10.1016/j.nec.2005.07. 004 7. Li J, Li Y, Kong F, Zhang D, Zhang Y, Shen Y (2014) Adjacent segment degeneration after single-level anterior cervical decompression and fusion: disc space distraction and its impact on clinical outcomes. J Clin Neurosci. doi:10.1016/j.jocn.2014.08. 019 8. Katsumi K, Yamazaki A, Watanabe K, Ohashi M, Shoji H (2013) Analysis of C5 palsy after cervical open-door laminoplasty: relationship between C5 palsy and foraminal stenosis. J Spinal Disord Tech 26:177–182. doi:10.1097/BSD. 0b013e31823db346 9. Gu Y, Cao P, Gao R, Tian Y, Liang L, Wang C, Yang L, Yuan W (2014) Incidence and risk factors of C5 palsy following posterior cervical decompression: a systematic review. PLoS One 9:e101933. doi:10.1371/journal.pone.0101933 10. Hosono N, Yonenobu K, Ono K (1996) Neck and shoulder pain after laminoplasty. A noticeable complication. Spine (Phila Pa 1976) 21:1969–1973 11. Lam FC, Irwin BJ, Poskitt KJ, Steinbok P (2009) Cervical spine instability following cervical laminectomies for Chiari II malformation: a retrospective cohort study. Childs Nerv Syst 25:71–76. doi:10.1007/s00381-008-0694-5 12. Yasuoka S, Peterson HA, MacCarty CS (1982) Incidence of spinal column deformity after multilevel laminectomy in children and adults. J Neurosurg 57:441–445. doi:10.3171/jns.1982.57.4. 0441 13. Guigui P, Benoist M, Deburge A (1998) Spinal deformity and instability after multilevel cervical laminectomy for spondylotic myelopathy. Spine (Phila Pa 1976) 23:440–447

14. Liu T, Yang HL, Xu YZ, Qi RF, Guan HQ (2011) ACDF with the PCB cage-plate system versus laminoplasty for multilevel cervical spondylotic myelopathy. J Spinal Disord Tech 24:213–220. doi:10.1097/BSD.0b013e3181e9f294 15. Ghogawala Z, Martin B, Benzel EC, Dziura J, Magge SN, Abbed KM, Bisson EF, Shahid J, Coumans JCE, Choudhri TF, Steinmetz MP, Krishnaney AA, King JT, Butler WE, Barker FG, Heary RF (2011) Comparative effectiveness of ventral vs dorsal surgery for cervical spondylotic myelopathy. Neurosurgery 68:622–631. doi:10.1227/NEU.0b013e31820777cf 16. Kristof RA, Kiefer T, Thudium M, Ringel F, Stoffel M, Kovacs A, Mueller C (2009) Comparison of ventral corpectomy and plate-screw-instrumented fusion with dorsal laminectomy and rod-screw-instrumented fusion for treatment of at least two vertebral-level spondylotic cervical myelopathy. Eur Spine J 18:1951–1956. doi:10.1007/s00586-009-1110-x 17. Charles C, Edwards IJGH (2002) Corpectomies versus laminoplasty for multilevel cervical myelopathy. Spine J 11:1168–1175. doi:10.1016/S1529-9430(01)00213-3 18. Shibuya S, Komatsubara S, Oka S, Kanda Y, Arima N, Yamamoto T (2009) Differences between subtotal corpectomy and laminoplasty for cervical spondylotic myelopathy. Spinal Cord 48:214–220. doi:10.1038/sc.2009.114 19. Yonenobu K, Hosono N, Iwasaki M, Asano M, Ono K (1992) Laminoplasty versus subtotal corpectomy. A comparative study of results in multisegmental cervical spondylotic myelopathy. Spine (Phila Pa 1976) 17:1281–1284 20. Hirai T, Okawa A, Arai Y, Takahashi M, Kawabata S, Kato T, Enomoto M, Tomizawa S, Sakai K, TorigoeI, Shinomiya K (2011) Middle-term results of a prospective comparative study of anterior decompression with fusion and posterior decompression with laminoplasty for the treatment of cervical spondylotic myelopathy. Spine (Phila Pa 1976) 23:1940–1947 21. Wada E, Suzuki S, Kanazawa A, Matsuoka T, Miyamoto S, Yonenobu K (2001) Subtotal corpectomy versus laminoplasty for multilevel cervical spondylotic myelopathy: a long-term followup study over 10 years. Spine (Phila Pa 1976) 26:1443–1447 (1448) 22. Seng C, Tow BPB, Siddiqui MA, Srivastava A, Wang L, Yew AKS, Yeo W, Khoo SHR, Balakrishnan NMS, Bin Abd Razak HR, Chen JLT, Guo CM, Tan SB, Yue W (2013) Surgically treated cervical myelopathy: a functional outcome comparison study between multilevel anterior cervical decompression fusion with instrumentation and posterior laminoplasty. Spine J 13:723–731. doi:10.1016/j.spinee.2013.02.038 23. Fehlings MG, Barry S, Kopjar B, Yoon ST, Arnold P, Massicotte EM, Vaccaro A, Brodke DS, Shaffrey C, Smith JS, Woodard E, Banco RJ, Chapman J, Janssen M, Bono C, Sasso R, Dekutoski M, Gokaslan ZL (2013) Anterior versus posterior surgical approaches to treat cervical spondylotic myelopathy: outcomes of the prospective multicenter AOSpine North America CSM study in 264 patients. Spine (Phila Pa 1976) 38:2247–2252 24. Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. Newcastle–Ottawa quality assessment scale for case control studies and cohort studies. www.ohri.ca/programs/clin ical_epidemiology/oxford.htm. Accessed 5 April 2014 25. Zhu B, Xu Y, Liu X, Liu Z, Dang G (2013) Anterior approach versus posterior approach for the treatment of multilevel cervical spondylotic myelopathy: a systemic review and meta-analysis. Eur Spine J 22:1583–1593. doi:10.1007/s00586-013-2817-2 26. Liu X, Wang H, Zhou Z, Jin A (2014) Anterior decompression and fusion versus posterior laminoplasty for multilevel cervical compressive myelopathy. Orthopedics 37:e117–e122. doi:10. 3928/01477447-20140124-12 27. Lawrence BD, Jacobs WB, Norvell DC, Hermsmeyer JT, Chapman JR, Brodke DS (2013) Anterior versus posterior approach for

123

Eur Spine J treatment of cervical spondylotic myelopathy a systematic review. Spine (Phila Pa 1976). 38(22 Suppl 1):S173–S182 28. Liu X, Min S, Zhang H, Zhou Z, Wang H, Jin A (2014) Anterior corpectomy versus posterior laminoplasty for multilevel cervical myelopathy: a systematic review and metaanalysis. Eur Spine J 23:362–372. doi:10.1007/s00586-0133043-7

123

29. Liu T, Xu W, Cheng T, Yang H (2011) Anterior versus posterior surgery for multilevel cervical myelopathy, which one is better? A systematic review. Eur Spine J 20(2):224–235 30. Xu J, Zhang K, Ma X, Yin Q, Wu Z, Xia H, Wang Z (2011) Systematic review of cohort studies comparing surgical treatment for multilevel ossification of posterior longitudinal ligament anterior vs posterior approach. Orthopedics 34(8):e397–e402

Comparison of anterior approach versus posterior approach for the treatment of multilevel cervical spondylotic myelopathy.

The purpose of this study is to evaluate the clinical outcomes, complications, and surgical trauma between anterior and posterior approaches for the t...
452KB Sizes 0 Downloads 11 Views