- Email: [email protected]
Treatment of Atlantoaxial Tuberculosis with Neurological Impairment: A Systematic Review
Journal Pre-proof Treatment of atlantoaxial tuberculosis with neurological impairment: A systematic review DingSu Bao, MD, Lang Li, MD, Min Gong, MD, Zhou Xiang, MD PII:
S1878-8750(19)32503-3
DOI:
https://doi.org/10.1016/j.wneu.2019.09.073
Reference:
WNEU 13372
To appear in:
World Neurosurgery
Received Date: 7 April 2019 Revised Date:
12 September 2019
Accepted Date: 13 September 2019
Please cite this article as: Bao D, Li L, Gong M, Xiang Z, Treatment of atlantoaxial tuberculosis with neurological impairment: A systematic review, World Neurosurgery (2019), doi: https://doi.org/10.1016/ j.wneu.2019.09.073. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
Treatment of atlantoaxial tuberculosis with neurological impairment: A systematic review
DingSu Bao 1,2, MD
Lang Li 1,MD
Min Gong 1, MD • Zhou Xiang 1, MD
Investigation performed at the Department of Orthopaedic Surgery, West China Hospital, Sichuan University, Chengdu, China
1. Department of Orthopaedics, West China Hospital, Sichuan University, Chengdu, China 2. Hospital Traditional Chinese Medicine Affiliated to Southwest Medical University, Luzhou, Sichuan, China
Address correspondence to: to: Dr. Zhou Xiang, Department of Orthopaedics, West China Hospital, Sichuan University, No. 37 Guoxue Xiang, Chengdu 610041, Sichuan, China. Tel: +86Fax: +86E+86-18980601393; +86-02885423438; E-mail: [email protected]
Lang Li authors contributed contributed equally to this work. Key Words: Atlantoaxial joint, craniovertebral junction, craniovertebral junction, treatment, surgery, conservation.
Background: Tuberculosis (TB) in the atlantoaxial joint is a rare disease. However, the treatment of atlantoaxial TB with neurological impairment is controversial. The aim of this review is to provide clinical outcomes of surgical and non-surgical management of atlantoaxial TB. Methods: Databases, including PubMed, Embase and the Cochrane Central Register of Controlled Trials, were searched for English literature describing the treatment of atlantoaxial TB with neurological deficits. The outcomes of conservative and surgical treatment approaches, including treatment failure, death, changes in neurological impairment and complications, were compared by performing odds ratio (OR) analysis. Results: Overall, 24 studies (247 patients) meeting the inclusion criteria were analysed. There were 94 (38%) patients treated conservatively and 153 (62%) patients treated surgically. The rate of poor outcomes was greater in the conservative group (14.89%) than in the surgery group (1.3%) [OR, 0.081; 95% confidence interval (CI), 0.016–0.39]. There was no significant difference in mortality between the conservative (1.06%) and surgery (3.27%) groups (OR, 3.28; 95% CI, 0.494–27.381). There was no significant difference in muscle power improvement between the two treatments (conservative: 95.7%; surgery: 94.8%; OR, 1.353; 95% CI, 0.291–4.925). Conclusion: Conservative and surgical treatments both significantly improved neurological deficits in most of the patients. In comparison with conservative treatment, surgical treatment reduced treatment failures without significantly increasing the rates of neurological deficit improvement or mortality. Key Words: Atlantoaxial joint, craniovertebral junction, craniovertebral junction, treatment, surgery, conservation.
Tuberculosis (TB) in the spine is an uncommon disease that occurs in 1%–2% of all patients with TB [1]. TB in the atlantoaxial joint is especially rare and accounts for 0.3%–4.8% of all spinal TB cases [2-4]. Hilton [5] was the first to report a case of atlantoaxial TB. Owing to the destruction of ligament and bone of atlantoaxis, atlantoaxial TB can lead to instability of the joint, which results in neck pain, torticollis, neurological impairment (spastic quadriparesis/monoparesis/hemiparesis, respiratory distress and bladder disturbance) and even sudden death [6]. Some reports have proposed classification and grade schemes for atlantoaxial TB to guide treatment. 1
Goel and Lifeso separately described three stages of atlantoaxial TB that were based on the pathogenesis of the atlantoaxial facet [7,8]. Teegala developed the atlantoaxial joint TB grading system comprising active neck movement restriction, motor power and radiological abnormalities [9]. A radiological staging system based on radiological findings has been proposed by Khandelwal [10]. Treatment for atlantoaxial TB can follow three approaches: conservative treatment, surgical treatment and a mixed approach based on clinical symptoms and radiological findings, which is currently preferred by more surgeons [11]. Patients without neurological symptoms and atlantoaxial instability usually accept conservative treatment. However, the treatment for patients with neurological impairment and atlantoaxial instability is controversial. Gupta speculated that surgery was not necessary for patients with advanced stages of disease [12]. In contrast, Arunkumar speculated that all patients should undergo surgery to achieve early immobilisation and rehabilitation [6]. Because atlantoaxial TB is uncommon, there is a lack of high-quality studies leading to a hardly established evidence-based algorithm available to support rational treatment. Atlantoaxial TB with neurological impairment is the most serious stage in patients. Therefore, the purpose of this systematic review was to answer the following questions: (1) What are the results of different treatments for atlantoaxial TB with neurological symptoms? (2) Which treatment approach (conservative vs. surgery) is more suitable for atlantoaxial TB with neurological impairment? (3) Which surgical methods for atlantoaxial TB are better?
Materials and Methods Search strategy This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [13]. We searched databases, including PubMed, Embase and the Cochrane Central Register of Controlled Trials from their establishment to August 2019 using the following terms: (suboccipital OR craniovertebral OR atlantoaxial) AND (Pott's disease OR tuberculosis). Only English-language studies with the full text available were included. When several publications reported findings for the same patients, the most recent or most complete study was chosen. Selection criteria Irrelevant articles were treated according to the flowchart shown in Figure 1. Studies were selected 2
according to the following criteria: (1) studies comprising patients with atlantoaxial TB as well as neurological impairment; (2) those comprising ≥2 patients; (3) those with an average follow-up time of ≥1 year; (4) those involving treatments such as conservative or surgical approaches; (5) those reporting ≥1 of the following outcomes: treatment failure (defined as conservative or surgical treatment failure requiring surgery or a second surgery, respectively), death, change in neurological impairment after treatment and complications and (6) full-text studies published in English. Letters, comments, abstracts for conferences, study protocols, reviews and animal studies were excluded. Data extraction First, two authors (Min Gong and Guo Ming Liu) independently reviewed all titles, abstracts and full texts to determine eligible articles. They reached agreement by discussion. If they had a dispute, they consulted the third author (Zhou Xiang) to finally make a decision. Second, the following information was extracted: name of first author, year of publication, study design, country location, demographic information (age and sex), number of patients, grade of neurological impairment, treatment approach, anti-TB treatment time and follow-up period and treatment outcomes, including treatment failure, deaths, change in neurological impairment after treatment and complications. Statistical analysis Treatment failure (defined as the need for surgery in patients who received conservative treatment and the need for revision surgery for patients who had undergone surgery), death, changes in neurological impairment after treatment (including improvement in muscle power, no improvement in sphincter disturbances and no improvement in respiratory distress) and complications were assessed by determining the 95% confidence intervals (CIs) and compared between the conservative and surgical treatment methods using odds ratio (OR) analysis. SPSS statistics 22 software was used for all statistical analyses.
Results After searching the database, 281 studies were chosen. Following the PRISMA guideline, we finally selected 24 studies meeting the inclusion criteria and obtained the full reports (Fig. 1) [2,4,6,8,9,11,12,14-30]. There were two level III studies and twenty two level IV studies. The studies included a total of 247 patients (age range, 3–85 years) with atlantoaxial TB and neurological 3
impairments, including motor weakness, respiratory distress and sphincter disturbances. There were 94 (38%) patients treated conservatively and 153 (62%) patients treated surgically. There were 27 (11%) patients with sphincter disturbances, and five of them in one study did not show improvement [14]. There were 25 (10%) patients with respiratory distress. Patients with sphincter disturbances and respiratory distress could not be included in the statistical analysis because some of the studies did not report the outcomes after treatment (Table 1). The conservative treatment for 46 (49%) patients with instability involved first placing them in skeletal traction after reduction or stability brace fixation, including the fixation of a Halo jacket or hard collar. The 48 (51%) patients without instability only received stability brace fixation using a Halo jacket or hard collar. In the studies included, surgical measures, such as anterior decompression (transoral drainage, transoral approach or transcervical retropharyngeal approach for both odontoidectomy and C1-2 decompression) and/or posterior fixation (occipitocervical fusion and C1-2 fixation with wires or screws), were adopted. Sixty-two patients were only treated with posterior fixation and 89 patients were treated with combined anterior decompression and posterior fixation. Only four patients underwent a single anterior surgical approach (Table 1). Treatment failure was higher in the conservative group (14.89%; 95% CI, 8.39%–23.72%) than in the surgery group (1.3%; 95% CI, 0.16%–4.7%) (OR, 0.081; 95% CI, 0.016–0.39). The most common reason for conservative treatment failure was mobile atlantoaxial dissociation (AAD) (71.4%). One patient in the conservative group died of acute hydrocephalus secondary to TB. There were five deaths, including those due to fulminant tubercular meningitis (n = 2), severe respiratory compromise (n = 1), unstable atlantoaxial joint (n = 1) and systemic problems (n = 1), in the surgical group. There was no significant difference in the mortality between the conservative (1.06%; 95% CI, 0.3%–5.78%) and surgery (3.27%; 95% CI, 1.08%–7.56%) groups (OR, 3.28; 95% CI, 0.494–27.381). There was also no significant difference in the muscle power improvement between the two treatments (conservative: 95.7%; 95% CI, 89.46%–98.83%; surgery: 94.8%; 95% CI, 91.55%–98.53%) (OR, 1.29; 95% CI, 0.291–4.925). There were 12 complications (12.7%) reported in the conservative group and 12 complications (7.8%) in the surgical group. Statistical analysis of the individual complications was not completed because of the lack of consistency among the studies. The raw data for complications are presented in Table 2. There were 89 patients treated by both the anterior approach for decompression and the posterior 4
approach for fusion plus fixation, and 62 patients were only treated by the posterior approach. There were no significant differences in treatment failure, deaths and muscle power improvement between the two treatment approaches (P > 0.05) (Table 3).
Discussion We finished a thorough search of databases and carefully reviewed the studies meeting the selection criteria. This systematic review found that for patients with neurological impairment caused by atlantoaxial TB, in comparison with conservative treatment, surgical treatment was associated with a significantly lower treatment failure rate without increasing the mortality rate. There was no significant difference in muscle power improvement between the surgical and conservative groups. There was also no significant difference in treatment results between the anterior approach for decompression plus the posterior approach and the posterior approach only. This review of the results from the limited number of English-language studies that have been conducted provides an assessment of the available evidence that may be useful for selecting the best treatment for rare atlantoaxial TB. Atlantoaxial TB is often secondary to the other site of the body, as first it leads to bone destruction, then involves the ligament and finally causes instability and neurological deficit [24]. The progression of atlantoaxial TB was divided into three stages by Goel. In stage 1, only the unilateral cancellous facet of atlas is involved without instability and neurological impairment. In stage 2, the unilateral bone damage progresses to impairment, which can affect the atlantoaxial joint complex and cause rotatory dislocation, but it is seldom accompanied by neurological deficit. In stage 3, the bilateral atlantoaxial joint is involved and usually causes severe instability with neurological deficit. Therefore, most patients with neurological signs have an unstable atlantoaxial joint [7]. Behari speculated that effective antitubercular therapy may produce a fibrous reaction and lead to stabilisation of the atlantoaxial joint [25]. Gupta reported that 11 patients with stage III were treated conservatively following a protocol that included skeletal traction for 3–10 days and then a Halo jacket for 3 months. He questioned the necessity of surgery because his study found that non-surgical intervention achieved good results, such as full clinical recovery and a stable atlantoaxial joint resulting from regrowth of bony structures [12]. This opinion was also advocated by authors of other studies [4,19,23,26,27,29]. In contrast, some surgeons preferred surgery for all patients with unstable atlantoaxial joints to reconstruct the stability [6,13,15,16,18,22,30]. In this review, we found that conservative treatment for 14 patients (15%) failed 5
and required to be followed by surgical intervention (10 patients with mobile AAD). Furthermore, six patients without treatment failure had a complication of fixed atlantoaxial dislocation in the conservative group. The outcomes for these patients suggest that it is difficult to recover the stability of an atlantoaxial joint with serious destruction of bone by conservative treatment. Therefore, we recommend surgery for patients with both AAD and neurological deficit. Neurological deficits, including paralysis, bladder disturbance and respiratory distress, caused by atlantoaxial TB occur only when the spinal cord is compressed by instability, deformity, basilar invagination or paravertebral or epidural abscesses at the atlantoaxial joint. Hoffman reported that >60% compression of the spinal canal can cause a neurological deficit [31]. It is necessary to decompress the compression by using some methods in time to improve neurological deficits. Some studies have shown that cervical tong traction reduced atlantoaxial dislocation, deformity and abscesses, which resulted in decompression of the cord [12,19,26-28]. Other studies have reported that surgery, especially via the anterior approach, is necessary to effectively decompress the spinal cord in a timely manner [6,15,18]. This review found no significant difference in improving neurological deficits between the surgical group and conservative group or between posterior fixation and the combination of anterior decompression with posterior fixation. Both conservative treatment and surgical treatment improved neurological deficits in 95% of the patients. These findings were obtained because of the following: 1) the ratio of the cervical cord/atlantoaxial canal diameter is lower than the subaxial cervical spine diameter and 2) antitubercular medicine is more effective because of the rich blood supply in the oropharynx. However, we still speculated that surgery had some of the following advantages: First, surgery quickly reduced dislocation caused by atlantoaxial TB, which led to the patients needing shorter hospitalisation and faster recovery than those associated with the conservative treatment. Second, compression of huge anterior abscesses can be effectively and quickly debrided by using the anterior approach. Although huge abscesses can be reduced by adequate antitubercular chemotherapy alone [32], it requires a long time and results in no improvement in neurological deficits [15]. Third, basilar invagination and irreducible AAD need thorough anterior decompression and posterior fixation, which can more often provide a more stable atlantoaxial joint than that observed with conservative treatment [6,15,21]. Arora speculated that the there was a high rate of serious complications associated with surgery in case of atlantoaxial TB. However, there were 12 (7.95%) complications in the surgery group in this 6
review. Previously, Fang reported anterior depression and arthrodesis for atlantoaxial TB with neurological deficit in 1985, which resulted in a 50% graft failure rate due to no fixation. In 2003, Sinha and Behari each used the fusion technique procedure of Brooks and experienced only one patient with fracture of the C1 arch (that required revision surgery) and another patient with wire breakage (that did not require revision surgery). These complications, however, can be avoided by using modern fixation and fusion, such as occipitocervical screw fixation fusion, C1-2 transarticular screw fixations and C1-2 screw fixations. The anterior approach for C1-2, such as transoral decompression, transoral resection of the odontoid and transcervical retropharyngeal decompression, has some complications, including recurrent laryngeal nerve injury, oesophageal injury, tracheal injury, major vessel injury and oral wound problems. In this review, complications associated with the anterior approach were low and included transient laryngeal nerve palsy (1.3%), transient dysphagia (1.3%) and oral wound dehiscence (0.7%). Oral wound infection is a serious complication. Patients with atlantoaxial TB theoretically have higher rates of oral wound infection than patients with non-infectious diseases. However, this review comprised no patients with oral wound infections and only one patient with oral wound dehiscence. Some new technologies have been used to treat atlantoaxial TB, aiming to reduce the incidence of complications. Yadav reported that the endoscopic transoral approach was safe and effective and allowed early oral feeding without palatal splitting and tongue oedema [33]. Other studies have reported that posterior debridement and fixation fusion methods were effective for anterior decompression, which avoided the anterior approach complications [17,34]. Overall, we speculated that the rate of surgical complications was relatively low. Death (particularly sudden death) can occur in patients with atlantoaxial TB caused by unrecognised instability and other system TB. In this review, the mortality rate was very low (2.45%). The risk factors of mortality do not include the treatment approach; however, they do include respiratory distress and secondary cerebral TB. There were several limitations of this review. First, 23 of the included studies were retrospective case series and only two studies were prospective. Therefore, there were unavoidable inherent shortcomings that are present in any systematic review. We recommended the level generated by this bibliography
was III. Second, the number of patients involved in the review was small and short of individual patient data, causing incomplete subgroup analysis on the grade of neurological deficit and age of patients. Though more high quality and large sample studies will be needed, it is hardly finished 7
because of scarce of the atlantoaxial TB. Third, TB is common in some poor areas and developing countries, leading to the exclusion of non-English literature. In conclusion, both conservative and surgical treatment significantly improved neurological deficits caused by atlantoaxial TB. The rate of treatment failure was lower for surgical treatment than for conservative treatment and the rates of neurological deficit and mortality did not increase. Anterior decompression plus posterior fusion fixation was not superior to the posterior approach with respect to treatment failure, death and neurological deficit improvement. Surgeons should choose the treatment approach for atlantoaxial TB primarily according to the personality of the patients.
Funding
This study was funded by Chinese-German cooperation: Repair and Reconstruction of
Musculoskeletal Tissue (GZ 1219) Compliance with ethical standards Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
Ethical approval is not applicable as this is a systematic review.
Informed consent
The nature of this article does not require informed consent.
References 1.
Jutte PC and Van Loenhout-Rooyackers JH (2006) Routine surgery in addition to chemotherapy for treating spinal tuberculosis. Cochrane Database Syst Rev 5:CD004532
2.
Lal AP, Rajshekhar V, Chandy MJ (1992) Management strategies in tuberculous atlanto-axial dislocation.Br J Neurosurg 6(6):529-535
3.
Edwards RJ, David KM, Crockard HA (2000) Management of tuberculomas of the craniovertebral junction. Br J Neurosurg 14(1):19-22
4.
Qureshi MA1, Afzal W, Khalique AB, Pasha IF, Aebi M (2013) Tuberculosis of the craniovertebral junction. Eur Spine J. 22 Suppl 4:612-7. doi: 10.1007/s00586-012-2497-3. Epub 2012 Oct 5
5.
Pandya SK (1970) John Hilton’s contributions on atlanto-axial disease – A forgotten chapter in the history of neurosurgery. Neurol India 18(3):147-57
6.
Arunkumar MJ, Rajshekhar V (2002) Outcome in neurologically impaired patients with craniovertebral junction tuberculosis: Results of combined anteroposterior surgery. J Neurosurg 97 2 Suppl: 166-71 8
7.
Goel A (2016) Tuberculosis of craniovertebral junction: Role of facets in pathogenesis and treatment. J Craniovertebr Junction Spine 7(3):129-30
8.
Lifeso R (1987) Atlanto-axial tuberculosis in adults. J Bone Joint Surg Br 69(2):183-7
9.
Teegala R, Kumar P, Kale S S (2008) Craniovertebral junction tuberculosis: a new comprehensive therapeutic strategy[J]. Neurosurgery 63(5):946-955
10. Khandelwal N, Khosla V K, Malik N (1991) Tuberculous atlanto-axial dislocation [M]// Proceedings of the XIV Symposium Neuroradiologicum. Springer Berlin Heidelberg, pp:106-108 11. Sridharan S, Arumugam T (2017) Atlantoaxial tuberculosis: Outcome analysis. Int J Mycobacteriol 6(2):127-131. 12. Mukherjee K, Gupta S, Mohindra S (2007) Craniovertebral Junction Tuberculosis: Is Surgery Required? Skull Base, Suppl 2 13.
Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6(7):e1000097.
14. Shukla D, Mongia S, Devi B I(2005) Management of craniovertebral junction tuberculosis. Surgical Neurology 63(2):101-106 15.
Xing S, Gao Y, Gao K (2015) Anterior Cervical Retropharyngeal Debridement Combined with Occipital Cervical Fusion to Upper Cervical Tuberculosis. Spine 41(2):104-110
16. Chatterjee S, Das A (2015) Craniovertebral tuberculosis in children: experience of 23 cases and proposal for a new classification. Childs Nervous System 31(8):1341-1345. 17. Zhang H, Sheng B, Tang M (2013) One-stage surgical treatment for upper thoracic spinal tuberculosis by internal fixation, debridement, and combined interbody and posterior fusion via posterior-only approach. European spine journal 22(3):616-623. 18.
Chaudhary K, Potdar P, Bapat M (2012) Structural odontoid lesions in craniovertebral tuberculosis: a review of 15 cases. Spine 37(14):E836
19. Arora S, Sabat D, Maini L(2011) The results of nonoperative treatment of craniovertebral junction tuberculosis: a review of twenty-six cases.Journal of Bone & Joint Surgery-american Volume 93(6):540-547. 20. Kotil K, Kalayci M, Bilge T (2007) Management of cervicomedullary compression in patients with congenital and acquired osseous-ligamentous pathologies. Journal of Clinical Neuroscience, 14(6):540-549 21.
Kalra S K, Kumar R, Mahapatra A K (2007) Tubercular atlantoaxial dislocation in children: an institutional experience. Journal of Neurosurgery 107(2 Suppl):111-118
22. Bapat M R, Lahiri V J, Harshavardhan N S (2007) Role of transarticular screw fixation in tuberculous atlanto-axial instability. European Spine Journal 16(2):187-197 23. Chadha M, Agarwal A, Singh A P (2007) Craniovertebral tuberculosis: a retrospective review of 13 cases managed conservatively. Spine 32(15):1629-34 24. Sinha S, Singh A K, Gupta V (2003) Surgical management and outcome of tuberculous atlantoaxial dislocation: a 15-year experience. Neurosurgery 52(2):331-338 25. Behari S, Nayak S R, Bhargava V (2003) Craniocervical tuberculosis: protocol of surgical management. Neurosurgery 52(1):72 26. Krishnan A, Patkar D, Patankar T (2001) Craniovertebral junction tuberculosis: a review of 29 cases. Journal of Computer Assisted Tomography 25(2):171 9
27. Dhammi I, Singh S, Jain A (2001) Hemiplegic/monoplegic presentation of cervical spine (C1-C2) tuberculosis. European Spine Journal 10(6):540-544 28. Tuli S M (1974) Tuberculosis of the craniovertebral region. Clinical Orthopaedics & Related Research 104(104):209. 29. Akhaddar A, Gourinda H, Gazzaz M, Elmadhi T, Elalami Z, MiriA (1999) Craniocervical junction tuberculosis in children. Rev Rhum Engl Ed 66(12):739–742 30. Biao Wang, Rongan Shang, Tong Yang, Haiping Zhang, Huimin Hu, Wei Hu ,Dingjun Hao (2019)Evaluation of clinical outcomes of one-stage anterior and posterior surgical treatment for atlantoaxial tuberculosis complicated with neurological
damage.
BMC
Musculoskelet Disord. 2019 Apr 6;20(1):148. 31. Hoffman EB, Crosier JH, Cremin BJ (1993) Imaging in children with spinal tuberculosis.A comparison of radiography, computed tomography and magnetic resonance imaging. J Bone Joint Surg Br 75(2):233-239 32. Jain AK, Jena A, Dhammi IK (2000) Correlation of clinical course with magnetic resonance imaging in tuberculous myelopathy. Neurol India 48(2):132-9 33. Yadav Y R, Madhariya S N, Parihar V S (2013) Endoscopic transoral excision of odontoid process in irreducible atlantoaxial dislocation: our experience of 34 patients. J Neurol Surg A Cent Eur Neurosurg 74(3):162-167 34. Mandavia R, Fox R, Meir A (2017) Atlantoaxial TB with paralysis: posterior-only cervical approach with good results. JRSM Open 8(6): 2054270417697866
10
Table 1 Characteristic of the studies utilized for systematic review Author
nation
Age
Number of patients （F/M） ）
level
neurological signs before treatment
Wang2019[30]
China
19-46y
12(5/7)
IV
Sridharan[11] 2017
India
14-33y
5（ （3/2） ）
IV
ASIA: C (n=4), D (n=8) paralysis (n=5) RD (n=3) BD (n=3)
Xing[15] 2016
China
40±9y
11(4/7)
IV
JOA score: 8 (7-11)
Chatterjee[16] 2015 Qureshi[4] 2013 Zhang[17] 2013
India
4-14y
5
IV
Pakista n China
15-85y
12(5/7)
IV
6-14y
9(4/5)
IV
paresis (n=5) RD（ （n=1） ） ASIA: C (n=5), D (n=7) ASIA: B (n=1), C (n=3), D(n=5)
Chaudhary[18 ] 2012
India
14-62y
8(6/2)
IV
Arora[19] 2011
India
28.3 y (8-49)
17
IV
Teegala [9] 2008
India
3-60y
35
III
Kotil [20] 2007 Kalra[21] 2007
Turkey
14-72y
4(2/2)
IV
india
5-35y
13
Bapat[22] 2007 Chadha[23] 2007
India
10-47y
India
5-35y
Treatment
Anti-TB treatment
Follow-time
neurological signs after treatment
Death/ treatment failure
TRD+ OCF
18m
6.5 ±2.9y
ASIA: E(12)
0
C: collar(n=2) S: TORO +OCF (n=1); C1-2 fixation (n=1); OCF (n=1). ACRPD+ OCFPS(n=5) ACRPD+ OCFLSPS (n=6) TOD + PF (n=1); PF (n=4) CTT (n=3); collar (n=9)
18 m
9-29m
0/2(C)
18m
18m
C: 2 worse S: Improved RD： ：* BD： ：* Seven excellent/ two good/2 remained
*
2.4 y (1.8 -5y)
0/0
12m
3y（ （1-6y） ）
All improved RD： ：* 11 recovered fully/1 remained. 7 improved/2 remained
PF(n=9)
18 m
28.1 ±10.5m
0/0
0/5(C) 0/0
Frankel Grade: B (n=2), C (n=3), D (n=3); RD (n=2); BD (n=3). Lorenzo grade : II (n=5), III (n=8), IV (n=4) RD:3,BD:5 paralysis (n=34) RD (n= 7) BD (n=4)
OCF(n=4)， ，TAS + C1–2 wiring (n=2)， ，C1–2 wiring +C1 arch excision(n=2)
9-12 m
3.6 y(1.5-8y） ）
All recovered fully, RD: *, BD： ：normal
0/0
CTT
14m
25.2m
All improved RD:* BD: *
0/0
C : collar (n=27) S: TOD +OCF (n=8)
21 m
18.5m(8-40m） ）
0/5(C)
TOD+OCF (n=4)
*
12-45 m
1(s)/0
IV
Frankel Grade (n=4) Lorenzo grade
All improved RD: improved BD: normal All recovered fully
13
III
Nurick’s grade (1-3)
3
IV
Muscle power grade 4 (n=4)
D
irreducible AAD: AD+ PF 18m (n=9), reducible AAD: PF (n=4) TAS/OCF 12 m
21.53±21.37 m
12 improved/1 worse
0/1(S)
2 y (2–4y )
0/0
CTT +brace (n=3)
43 m(16-65m)
10 improved 3 remained All improved
19 m
0/0
continue Author
nation
Age
Number of patients （F/M） ）
level
neurological signs before treatment
Treatment
Anti-TB treatment
Follow-time
neurological signs after Death/ treatment treatment failure
Gupta[12] 2006 Shukla[14] 2005
India
9-55y
11(4/7)
IV
>1y
1(C)/0
8-78y
16
IV
CTT +Halo jacket (n=11) TOD+ PF (n=10); PF (n=6)
18 m
India
muscle power Grade 0-4+(n=11), BD(n= 1) quadriparesis (n=16) BD(n=6)
Sinha[24] 2003 Behari[25] 2003
India
17-45y
IV
1(s)/0
10-60y
IV
Ranawat score IIIA (n=9), IIIB (n=9) Grade of neurological disabilities II(n= 8), III (n=1), IV(n= 9) RD (n=9) , BD (n= 4)
TRD + C1-2 wire
India
18 (8/10) 18
Arunkumar[6] 2002
India
17-55y
9(1/8)
IV
Nurick grade 3.4(range 1–5).
Krishnan[26] 2001
india
*
3
IV
neurologic deficit
Dhammi[27] 2001 Lal[2] 1992
India
20-45y
IV
india
23y
3 （2/1） ） 4(1/3)
IV
Muscle power grade 0-3(n=3), BD(n=1) Muscle power2-4
Lifeso[8] 1987 Fang [28] 1985
Audi Arabias china
14-65y
5
IV
Cord involvment
3-51y
4
IV
MRC grade I-IV
TOD+fusion: 3 PF:1
Tuli[29] 1974
india
*
8
IV
Neural complcations
CTT + brace
18 m
9.3m (1-36 ）
All recovered fully BD:* All improved BD: only 1 improved.
18 m
2 to 5 y
All recovered fully
0/0
C: Minerva jacket or 21 m hard collar (n =7). S: PF(n=7) /TOD + PF (n=4)
2.5y（6 m-7y） ） All improved. RD improvement. BD:*.
1(s)/1(s)
TOD + OCF (n=2), TORO + OCF(n=5), C1-2 decompression + OCF (n=2) CTT + bracing
18 m
18.8 m (7–46)
Nurick grade 2.3
0/0
*
2y
2 improved 1 remained
0/1(C)
22,48m
All complete recovery
0/1(C)
28m(3-48)
All complete recovery
1(s)/0
36m
Recovery
1(s)/0
15m
15m
recovery
0/0
*
1-7y
7 recover completely 1 partial recovery
0/0
CTT+ brace (n=3)
18 m
C: Minerva jacket 18m (n=1) S: drainage of RPA (n=1), PF+ minerva jacket (n=2) Surgery: OCF+Halo 15m
ASIA: Asia impairment Scale; TORO: Trans oral resection of odontoid; TOD: trans-oral decompression; TRD: transcervical retropharyngeal decompression; OCF: Occipitocervical fusion. RD: Respiratory distress; BD: Bladder disturbance; TAS: transarticular screws; *: Not reported; PF: posterior fixation. CTT: cervical tongs traction; ENAD: endonasal approach decompression;
ACRPD: anterior cervical and retropharyngeal debridement; OCFLS: occipital cervical fusion by laminar screw; OCFLSPS: occipital cervical fusion by laminar screw and pedicle screw
Table 2 the results of the conservation and the surgery. Method of treatment
No. Patients
Treatment Failure
Deaths
improving Muscle power
Complications
Conservation
94
14 (4 patient for non-improving neurological deficit, 10 patients for mobile AAD)
1
90
12 (pressure sores: 5, fixed atlantoaxial dislocation : 6 , sacral decubitus ulcer: 1)
surgery
153
2 (brooks technique with fracture of the C1 arch, neurological deterioration after ACDF)
5
145
12 (Anterior approach: wound infection: 2; Laryngeal nerve palsy: 2; transient dysphagia: 2; bone grafts displaced:2. Oral wound dehiscence: 1. Posterior approach: wire breakage: 1; fracture of the C1 arch: 1;；infection of the iliac bone graft donor site：1 )
Table 3 the results of the anterior decompression plus posterior fusion fixation and the posterior approach Procedure
No. Patients
Treatment Failure
Deaths
Non-Improving Muscle power
A+P P P value
89 62
1 1 P>0.05
3 2 P>0.05
3 5 P>0.05
A + P: anterior decompression plus posterior fusion fixation, P: posterior approach.
Records identified through database searching(n=281)： pubmed(n=130);EMBASE(n=151); Cochrane library(n=0)
Records after duplicates removed (n =141)
Records excluded according to title and abstract (n=86)
Full-text articles assessed for eligibility (n=57)
Studies included in quantitative analysis (n=24)
text articles excluded with reasons： No neurological defects =5 Insufficient data=4 Less than 2 patients=23 Overlapping cohort=1
Figure 1: Flow diagram showing the selection process
Abbreviations TB, tuberculosis; AAD, atlantoaxial dissociation; OR, odds ratio；PRISMA, Preferred Reporting Items for Systematic Review and Meta-analysis ; CI, confidence interval;
S1878-8750(19)32503-3
DOI:
https://doi.org/10.1016/j.wneu.2019.09.073
Reference:
WNEU 13372
To appear in:
World Neurosurgery
Received Date: 7 April 2019 Revised Date:
12 September 2019
Accepted Date: 13 September 2019
Please cite this article as: Bao D, Li L, Gong M, Xiang Z, Treatment of atlantoaxial tuberculosis with neurological impairment: A systematic review, World Neurosurgery (2019), doi: https://doi.org/10.1016/ j.wneu.2019.09.073. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
Treatment of atlantoaxial tuberculosis with neurological impairment: A systematic review
DingSu Bao 1,2, MD
Lang Li 1,MD
Min Gong 1, MD • Zhou Xiang 1, MD
Investigation performed at the Department of Orthopaedic Surgery, West China Hospital, Sichuan University, Chengdu, China
1. Department of Orthopaedics, West China Hospital, Sichuan University, Chengdu, China 2. Hospital Traditional Chinese Medicine Affiliated to Southwest Medical University, Luzhou, Sichuan, China
Address correspondence to: to: Dr. Zhou Xiang, Department of Orthopaedics, West China Hospital, Sichuan University, No. 37 Guoxue Xiang, Chengdu 610041, Sichuan, China. Tel: +86Fax: +86E+86-18980601393; +86-02885423438; E-mail: [email protected]
Lang Li authors contributed contributed equally to this work. Key Words: Atlantoaxial joint, craniovertebral junction, craniovertebral junction, treatment, surgery, conservation.
Background: Tuberculosis (TB) in the atlantoaxial joint is a rare disease. However, the treatment of atlantoaxial TB with neurological impairment is controversial. The aim of this review is to provide clinical outcomes of surgical and non-surgical management of atlantoaxial TB. Methods: Databases, including PubMed, Embase and the Cochrane Central Register of Controlled Trials, were searched for English literature describing the treatment of atlantoaxial TB with neurological deficits. The outcomes of conservative and surgical treatment approaches, including treatment failure, death, changes in neurological impairment and complications, were compared by performing odds ratio (OR) analysis. Results: Overall, 24 studies (247 patients) meeting the inclusion criteria were analysed. There were 94 (38%) patients treated conservatively and 153 (62%) patients treated surgically. The rate of poor outcomes was greater in the conservative group (14.89%) than in the surgery group (1.3%) [OR, 0.081; 95% confidence interval (CI), 0.016–0.39]. There was no significant difference in mortality between the conservative (1.06%) and surgery (3.27%) groups (OR, 3.28; 95% CI, 0.494–27.381). There was no significant difference in muscle power improvement between the two treatments (conservative: 95.7%; surgery: 94.8%; OR, 1.353; 95% CI, 0.291–4.925). Conclusion: Conservative and surgical treatments both significantly improved neurological deficits in most of the patients. In comparison with conservative treatment, surgical treatment reduced treatment failures without significantly increasing the rates of neurological deficit improvement or mortality. Key Words: Atlantoaxial joint, craniovertebral junction, craniovertebral junction, treatment, surgery, conservation.
Tuberculosis (TB) in the spine is an uncommon disease that occurs in 1%–2% of all patients with TB [1]. TB in the atlantoaxial joint is especially rare and accounts for 0.3%–4.8% of all spinal TB cases [2-4]. Hilton [5] was the first to report a case of atlantoaxial TB. Owing to the destruction of ligament and bone of atlantoaxis, atlantoaxial TB can lead to instability of the joint, which results in neck pain, torticollis, neurological impairment (spastic quadriparesis/monoparesis/hemiparesis, respiratory distress and bladder disturbance) and even sudden death [6]. Some reports have proposed classification and grade schemes for atlantoaxial TB to guide treatment. 1
Goel and Lifeso separately described three stages of atlantoaxial TB that were based on the pathogenesis of the atlantoaxial facet [7,8]. Teegala developed the atlantoaxial joint TB grading system comprising active neck movement restriction, motor power and radiological abnormalities [9]. A radiological staging system based on radiological findings has been proposed by Khandelwal [10]. Treatment for atlantoaxial TB can follow three approaches: conservative treatment, surgical treatment and a mixed approach based on clinical symptoms and radiological findings, which is currently preferred by more surgeons [11]. Patients without neurological symptoms and atlantoaxial instability usually accept conservative treatment. However, the treatment for patients with neurological impairment and atlantoaxial instability is controversial. Gupta speculated that surgery was not necessary for patients with advanced stages of disease [12]. In contrast, Arunkumar speculated that all patients should undergo surgery to achieve early immobilisation and rehabilitation [6]. Because atlantoaxial TB is uncommon, there is a lack of high-quality studies leading to a hardly established evidence-based algorithm available to support rational treatment. Atlantoaxial TB with neurological impairment is the most serious stage in patients. Therefore, the purpose of this systematic review was to answer the following questions: (1) What are the results of different treatments for atlantoaxial TB with neurological symptoms? (2) Which treatment approach (conservative vs. surgery) is more suitable for atlantoaxial TB with neurological impairment? (3) Which surgical methods for atlantoaxial TB are better?
Materials and Methods Search strategy This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [13]. We searched databases, including PubMed, Embase and the Cochrane Central Register of Controlled Trials from their establishment to August 2019 using the following terms: (suboccipital OR craniovertebral OR atlantoaxial) AND (Pott's disease OR tuberculosis). Only English-language studies with the full text available were included. When several publications reported findings for the same patients, the most recent or most complete study was chosen. Selection criteria Irrelevant articles were treated according to the flowchart shown in Figure 1. Studies were selected 2
according to the following criteria: (1) studies comprising patients with atlantoaxial TB as well as neurological impairment; (2) those comprising ≥2 patients; (3) those with an average follow-up time of ≥1 year; (4) those involving treatments such as conservative or surgical approaches; (5) those reporting ≥1 of the following outcomes: treatment failure (defined as conservative or surgical treatment failure requiring surgery or a second surgery, respectively), death, change in neurological impairment after treatment and complications and (6) full-text studies published in English. Letters, comments, abstracts for conferences, study protocols, reviews and animal studies were excluded. Data extraction First, two authors (Min Gong and Guo Ming Liu) independently reviewed all titles, abstracts and full texts to determine eligible articles. They reached agreement by discussion. If they had a dispute, they consulted the third author (Zhou Xiang) to finally make a decision. Second, the following information was extracted: name of first author, year of publication, study design, country location, demographic information (age and sex), number of patients, grade of neurological impairment, treatment approach, anti-TB treatment time and follow-up period and treatment outcomes, including treatment failure, deaths, change in neurological impairment after treatment and complications. Statistical analysis Treatment failure (defined as the need for surgery in patients who received conservative treatment and the need for revision surgery for patients who had undergone surgery), death, changes in neurological impairment after treatment (including improvement in muscle power, no improvement in sphincter disturbances and no improvement in respiratory distress) and complications were assessed by determining the 95% confidence intervals (CIs) and compared between the conservative and surgical treatment methods using odds ratio (OR) analysis. SPSS statistics 22 software was used for all statistical analyses.
Results After searching the database, 281 studies were chosen. Following the PRISMA guideline, we finally selected 24 studies meeting the inclusion criteria and obtained the full reports (Fig. 1) [2,4,6,8,9,11,12,14-30]. There were two level III studies and twenty two level IV studies. The studies included a total of 247 patients (age range, 3–85 years) with atlantoaxial TB and neurological 3
impairments, including motor weakness, respiratory distress and sphincter disturbances. There were 94 (38%) patients treated conservatively and 153 (62%) patients treated surgically. There were 27 (11%) patients with sphincter disturbances, and five of them in one study did not show improvement [14]. There were 25 (10%) patients with respiratory distress. Patients with sphincter disturbances and respiratory distress could not be included in the statistical analysis because some of the studies did not report the outcomes after treatment (Table 1). The conservative treatment for 46 (49%) patients with instability involved first placing them in skeletal traction after reduction or stability brace fixation, including the fixation of a Halo jacket or hard collar. The 48 (51%) patients without instability only received stability brace fixation using a Halo jacket or hard collar. In the studies included, surgical measures, such as anterior decompression (transoral drainage, transoral approach or transcervical retropharyngeal approach for both odontoidectomy and C1-2 decompression) and/or posterior fixation (occipitocervical fusion and C1-2 fixation with wires or screws), were adopted. Sixty-two patients were only treated with posterior fixation and 89 patients were treated with combined anterior decompression and posterior fixation. Only four patients underwent a single anterior surgical approach (Table 1). Treatment failure was higher in the conservative group (14.89%; 95% CI, 8.39%–23.72%) than in the surgery group (1.3%; 95% CI, 0.16%–4.7%) (OR, 0.081; 95% CI, 0.016–0.39). The most common reason for conservative treatment failure was mobile atlantoaxial dissociation (AAD) (71.4%). One patient in the conservative group died of acute hydrocephalus secondary to TB. There were five deaths, including those due to fulminant tubercular meningitis (n = 2), severe respiratory compromise (n = 1), unstable atlantoaxial joint (n = 1) and systemic problems (n = 1), in the surgical group. There was no significant difference in the mortality between the conservative (1.06%; 95% CI, 0.3%–5.78%) and surgery (3.27%; 95% CI, 1.08%–7.56%) groups (OR, 3.28; 95% CI, 0.494–27.381). There was also no significant difference in the muscle power improvement between the two treatments (conservative: 95.7%; 95% CI, 89.46%–98.83%; surgery: 94.8%; 95% CI, 91.55%–98.53%) (OR, 1.29; 95% CI, 0.291–4.925). There were 12 complications (12.7%) reported in the conservative group and 12 complications (7.8%) in the surgical group. Statistical analysis of the individual complications was not completed because of the lack of consistency among the studies. The raw data for complications are presented in Table 2. There were 89 patients treated by both the anterior approach for decompression and the posterior 4
approach for fusion plus fixation, and 62 patients were only treated by the posterior approach. There were no significant differences in treatment failure, deaths and muscle power improvement between the two treatment approaches (P > 0.05) (Table 3).
Discussion We finished a thorough search of databases and carefully reviewed the studies meeting the selection criteria. This systematic review found that for patients with neurological impairment caused by atlantoaxial TB, in comparison with conservative treatment, surgical treatment was associated with a significantly lower treatment failure rate without increasing the mortality rate. There was no significant difference in muscle power improvement between the surgical and conservative groups. There was also no significant difference in treatment results between the anterior approach for decompression plus the posterior approach and the posterior approach only. This review of the results from the limited number of English-language studies that have been conducted provides an assessment of the available evidence that may be useful for selecting the best treatment for rare atlantoaxial TB. Atlantoaxial TB is often secondary to the other site of the body, as first it leads to bone destruction, then involves the ligament and finally causes instability and neurological deficit [24]. The progression of atlantoaxial TB was divided into three stages by Goel. In stage 1, only the unilateral cancellous facet of atlas is involved without instability and neurological impairment. In stage 2, the unilateral bone damage progresses to impairment, which can affect the atlantoaxial joint complex and cause rotatory dislocation, but it is seldom accompanied by neurological deficit. In stage 3, the bilateral atlantoaxial joint is involved and usually causes severe instability with neurological deficit. Therefore, most patients with neurological signs have an unstable atlantoaxial joint [7]. Behari speculated that effective antitubercular therapy may produce a fibrous reaction and lead to stabilisation of the atlantoaxial joint [25]. Gupta reported that 11 patients with stage III were treated conservatively following a protocol that included skeletal traction for 3–10 days and then a Halo jacket for 3 months. He questioned the necessity of surgery because his study found that non-surgical intervention achieved good results, such as full clinical recovery and a stable atlantoaxial joint resulting from regrowth of bony structures [12]. This opinion was also advocated by authors of other studies [4,19,23,26,27,29]. In contrast, some surgeons preferred surgery for all patients with unstable atlantoaxial joints to reconstruct the stability [6,13,15,16,18,22,30]. In this review, we found that conservative treatment for 14 patients (15%) failed 5
and required to be followed by surgical intervention (10 patients with mobile AAD). Furthermore, six patients without treatment failure had a complication of fixed atlantoaxial dislocation in the conservative group. The outcomes for these patients suggest that it is difficult to recover the stability of an atlantoaxial joint with serious destruction of bone by conservative treatment. Therefore, we recommend surgery for patients with both AAD and neurological deficit. Neurological deficits, including paralysis, bladder disturbance and respiratory distress, caused by atlantoaxial TB occur only when the spinal cord is compressed by instability, deformity, basilar invagination or paravertebral or epidural abscesses at the atlantoaxial joint. Hoffman reported that >60% compression of the spinal canal can cause a neurological deficit [31]. It is necessary to decompress the compression by using some methods in time to improve neurological deficits. Some studies have shown that cervical tong traction reduced atlantoaxial dislocation, deformity and abscesses, which resulted in decompression of the cord [12,19,26-28]. Other studies have reported that surgery, especially via the anterior approach, is necessary to effectively decompress the spinal cord in a timely manner [6,15,18]. This review found no significant difference in improving neurological deficits between the surgical group and conservative group or between posterior fixation and the combination of anterior decompression with posterior fixation. Both conservative treatment and surgical treatment improved neurological deficits in 95% of the patients. These findings were obtained because of the following: 1) the ratio of the cervical cord/atlantoaxial canal diameter is lower than the subaxial cervical spine diameter and 2) antitubercular medicine is more effective because of the rich blood supply in the oropharynx. However, we still speculated that surgery had some of the following advantages: First, surgery quickly reduced dislocation caused by atlantoaxial TB, which led to the patients needing shorter hospitalisation and faster recovery than those associated with the conservative treatment. Second, compression of huge anterior abscesses can be effectively and quickly debrided by using the anterior approach. Although huge abscesses can be reduced by adequate antitubercular chemotherapy alone [32], it requires a long time and results in no improvement in neurological deficits [15]. Third, basilar invagination and irreducible AAD need thorough anterior decompression and posterior fixation, which can more often provide a more stable atlantoaxial joint than that observed with conservative treatment [6,15,21]. Arora speculated that the there was a high rate of serious complications associated with surgery in case of atlantoaxial TB. However, there were 12 (7.95%) complications in the surgery group in this 6
review. Previously, Fang reported anterior depression and arthrodesis for atlantoaxial TB with neurological deficit in 1985, which resulted in a 50% graft failure rate due to no fixation. In 2003, Sinha and Behari each used the fusion technique procedure of Brooks and experienced only one patient with fracture of the C1 arch (that required revision surgery) and another patient with wire breakage (that did not require revision surgery). These complications, however, can be avoided by using modern fixation and fusion, such as occipitocervical screw fixation fusion, C1-2 transarticular screw fixations and C1-2 screw fixations. The anterior approach for C1-2, such as transoral decompression, transoral resection of the odontoid and transcervical retropharyngeal decompression, has some complications, including recurrent laryngeal nerve injury, oesophageal injury, tracheal injury, major vessel injury and oral wound problems. In this review, complications associated with the anterior approach were low and included transient laryngeal nerve palsy (1.3%), transient dysphagia (1.3%) and oral wound dehiscence (0.7%). Oral wound infection is a serious complication. Patients with atlantoaxial TB theoretically have higher rates of oral wound infection than patients with non-infectious diseases. However, this review comprised no patients with oral wound infections and only one patient with oral wound dehiscence. Some new technologies have been used to treat atlantoaxial TB, aiming to reduce the incidence of complications. Yadav reported that the endoscopic transoral approach was safe and effective and allowed early oral feeding without palatal splitting and tongue oedema [33]. Other studies have reported that posterior debridement and fixation fusion methods were effective for anterior decompression, which avoided the anterior approach complications [17,34]. Overall, we speculated that the rate of surgical complications was relatively low. Death (particularly sudden death) can occur in patients with atlantoaxial TB caused by unrecognised instability and other system TB. In this review, the mortality rate was very low (2.45%). The risk factors of mortality do not include the treatment approach; however, they do include respiratory distress and secondary cerebral TB. There were several limitations of this review. First, 23 of the included studies were retrospective case series and only two studies were prospective. Therefore, there were unavoidable inherent shortcomings that are present in any systematic review. We recommended the level generated by this bibliography
was III. Second, the number of patients involved in the review was small and short of individual patient data, causing incomplete subgroup analysis on the grade of neurological deficit and age of patients. Though more high quality and large sample studies will be needed, it is hardly finished 7
because of scarce of the atlantoaxial TB. Third, TB is common in some poor areas and developing countries, leading to the exclusion of non-English literature. In conclusion, both conservative and surgical treatment significantly improved neurological deficits caused by atlantoaxial TB. The rate of treatment failure was lower for surgical treatment than for conservative treatment and the rates of neurological deficit and mortality did not increase. Anterior decompression plus posterior fusion fixation was not superior to the posterior approach with respect to treatment failure, death and neurological deficit improvement. Surgeons should choose the treatment approach for atlantoaxial TB primarily according to the personality of the patients.
Funding
This study was funded by Chinese-German cooperation: Repair and Reconstruction of
Musculoskeletal Tissue (GZ 1219) Compliance with ethical standards Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
Ethical approval is not applicable as this is a systematic review.
Informed consent
The nature of this article does not require informed consent.
References 1.
Jutte PC and Van Loenhout-Rooyackers JH (2006) Routine surgery in addition to chemotherapy for treating spinal tuberculosis. Cochrane Database Syst Rev 5:CD004532
2.
Lal AP, Rajshekhar V, Chandy MJ (1992) Management strategies in tuberculous atlanto-axial dislocation.Br J Neurosurg 6(6):529-535
3.
Edwards RJ, David KM, Crockard HA (2000) Management of tuberculomas of the craniovertebral junction. Br J Neurosurg 14(1):19-22
4.
Qureshi MA1, Afzal W, Khalique AB, Pasha IF, Aebi M (2013) Tuberculosis of the craniovertebral junction. Eur Spine J. 22 Suppl 4:612-7. doi: 10.1007/s00586-012-2497-3. Epub 2012 Oct 5
5.
Pandya SK (1970) John Hilton’s contributions on atlanto-axial disease – A forgotten chapter in the history of neurosurgery. Neurol India 18(3):147-57
6.
Arunkumar MJ, Rajshekhar V (2002) Outcome in neurologically impaired patients with craniovertebral junction tuberculosis: Results of combined anteroposterior surgery. J Neurosurg 97 2 Suppl: 166-71 8
7.
Goel A (2016) Tuberculosis of craniovertebral junction: Role of facets in pathogenesis and treatment. J Craniovertebr Junction Spine 7(3):129-30
8.
Lifeso R (1987) Atlanto-axial tuberculosis in adults. J Bone Joint Surg Br 69(2):183-7
9.
Teegala R, Kumar P, Kale S S (2008) Craniovertebral junction tuberculosis: a new comprehensive therapeutic strategy[J]. Neurosurgery 63(5):946-955
10. Khandelwal N, Khosla V K, Malik N (1991) Tuberculous atlanto-axial dislocation [M]// Proceedings of the XIV Symposium Neuroradiologicum. Springer Berlin Heidelberg, pp:106-108 11. Sridharan S, Arumugam T (2017) Atlantoaxial tuberculosis: Outcome analysis. Int J Mycobacteriol 6(2):127-131. 12. Mukherjee K, Gupta S, Mohindra S (2007) Craniovertebral Junction Tuberculosis: Is Surgery Required? Skull Base, Suppl 2 13.
Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6(7):e1000097.
14. Shukla D, Mongia S, Devi B I(2005) Management of craniovertebral junction tuberculosis. Surgical Neurology 63(2):101-106 15.
Xing S, Gao Y, Gao K (2015) Anterior Cervical Retropharyngeal Debridement Combined with Occipital Cervical Fusion to Upper Cervical Tuberculosis. Spine 41(2):104-110
16. Chatterjee S, Das A (2015) Craniovertebral tuberculosis in children: experience of 23 cases and proposal for a new classification. Childs Nervous System 31(8):1341-1345. 17. Zhang H, Sheng B, Tang M (2013) One-stage surgical treatment for upper thoracic spinal tuberculosis by internal fixation, debridement, and combined interbody and posterior fusion via posterior-only approach. European spine journal 22(3):616-623. 18.
Chaudhary K, Potdar P, Bapat M (2012) Structural odontoid lesions in craniovertebral tuberculosis: a review of 15 cases. Spine 37(14):E836
19. Arora S, Sabat D, Maini L(2011) The results of nonoperative treatment of craniovertebral junction tuberculosis: a review of twenty-six cases.Journal of Bone & Joint Surgery-american Volume 93(6):540-547. 20. Kotil K, Kalayci M, Bilge T (2007) Management of cervicomedullary compression in patients with congenital and acquired osseous-ligamentous pathologies. Journal of Clinical Neuroscience, 14(6):540-549 21.
Kalra S K, Kumar R, Mahapatra A K (2007) Tubercular atlantoaxial dislocation in children: an institutional experience. Journal of Neurosurgery 107(2 Suppl):111-118
22. Bapat M R, Lahiri V J, Harshavardhan N S (2007) Role of transarticular screw fixation in tuberculous atlanto-axial instability. European Spine Journal 16(2):187-197 23. Chadha M, Agarwal A, Singh A P (2007) Craniovertebral tuberculosis: a retrospective review of 13 cases managed conservatively. Spine 32(15):1629-34 24. Sinha S, Singh A K, Gupta V (2003) Surgical management and outcome of tuberculous atlantoaxial dislocation: a 15-year experience. Neurosurgery 52(2):331-338 25. Behari S, Nayak S R, Bhargava V (2003) Craniocervical tuberculosis: protocol of surgical management. Neurosurgery 52(1):72 26. Krishnan A, Patkar D, Patankar T (2001) Craniovertebral junction tuberculosis: a review of 29 cases. Journal of Computer Assisted Tomography 25(2):171 9
27. Dhammi I, Singh S, Jain A (2001) Hemiplegic/monoplegic presentation of cervical spine (C1-C2) tuberculosis. European Spine Journal 10(6):540-544 28. Tuli S M (1974) Tuberculosis of the craniovertebral region. Clinical Orthopaedics & Related Research 104(104):209. 29. Akhaddar A, Gourinda H, Gazzaz M, Elmadhi T, Elalami Z, MiriA (1999) Craniocervical junction tuberculosis in children. Rev Rhum Engl Ed 66(12):739–742 30. Biao Wang, Rongan Shang, Tong Yang, Haiping Zhang, Huimin Hu, Wei Hu ,Dingjun Hao (2019)Evaluation of clinical outcomes of one-stage anterior and posterior surgical treatment for atlantoaxial tuberculosis complicated with neurological
damage.
BMC
Musculoskelet Disord. 2019 Apr 6;20(1):148. 31. Hoffman EB, Crosier JH, Cremin BJ (1993) Imaging in children with spinal tuberculosis.A comparison of radiography, computed tomography and magnetic resonance imaging. J Bone Joint Surg Br 75(2):233-239 32. Jain AK, Jena A, Dhammi IK (2000) Correlation of clinical course with magnetic resonance imaging in tuberculous myelopathy. Neurol India 48(2):132-9 33. Yadav Y R, Madhariya S N, Parihar V S (2013) Endoscopic transoral excision of odontoid process in irreducible atlantoaxial dislocation: our experience of 34 patients. J Neurol Surg A Cent Eur Neurosurg 74(3):162-167 34. Mandavia R, Fox R, Meir A (2017) Atlantoaxial TB with paralysis: posterior-only cervical approach with good results. JRSM Open 8(6): 2054270417697866
10
Table 1 Characteristic of the studies utilized for systematic review Author
nation
Age
Number of patients （F/M） ）
level
neurological signs before treatment
Wang2019[30]
China
19-46y
12(5/7)
IV
Sridharan[11] 2017
India
14-33y
5（ （3/2） ）
IV
ASIA: C (n=4), D (n=8) paralysis (n=5) RD (n=3) BD (n=3)
Xing[15] 2016
China
40±9y
11(4/7)
IV
JOA score: 8 (7-11)
Chatterjee[16] 2015 Qureshi[4] 2013 Zhang[17] 2013
India
4-14y
5
IV
Pakista n China
15-85y
12(5/7)
IV
6-14y
9(4/5)
IV
paresis (n=5) RD（ （n=1） ） ASIA: C (n=5), D (n=7) ASIA: B (n=1), C (n=3), D(n=5)
Chaudhary[18 ] 2012
India
14-62y
8(6/2)
IV
Arora[19] 2011
India
28.3 y (8-49)
17
IV
Teegala [9] 2008
India
3-60y
35
III
Kotil [20] 2007 Kalra[21] 2007
Turkey
14-72y
4(2/2)
IV
india
5-35y
13
Bapat[22] 2007 Chadha[23] 2007
India
10-47y
India
5-35y
Treatment
Anti-TB treatment
Follow-time
neurological signs after treatment
Death/ treatment failure
TRD+ OCF
18m
6.5 ±2.9y
ASIA: E(12)
0
C: collar(n=2) S: TORO +OCF (n=1); C1-2 fixation (n=1); OCF (n=1). ACRPD+ OCFPS(n=5) ACRPD+ OCFLSPS (n=6) TOD + PF (n=1); PF (n=4) CTT (n=3); collar (n=9)
18 m
9-29m
0/2(C)
18m
18m
C: 2 worse S: Improved RD： ：* BD： ：* Seven excellent/ two good/2 remained
*
2.4 y (1.8 -5y)
0/0
12m
3y（ （1-6y） ）
All improved RD： ：* 11 recovered fully/1 remained. 7 improved/2 remained
PF(n=9)
18 m
28.1 ±10.5m
0/0
0/5(C) 0/0
Frankel Grade: B (n=2), C (n=3), D (n=3); RD (n=2); BD (n=3). Lorenzo grade : II (n=5), III (n=8), IV (n=4) RD:3,BD:5 paralysis (n=34) RD (n= 7) BD (n=4)
OCF(n=4)， ，TAS + C1–2 wiring (n=2)， ，C1–2 wiring +C1 arch excision(n=2)
9-12 m
3.6 y(1.5-8y） ）
All recovered fully, RD: *, BD： ：normal
0/0
CTT
14m
25.2m
All improved RD:* BD: *
0/0
C : collar (n=27) S: TOD +OCF (n=8)
21 m
18.5m(8-40m） ）
0/5(C)
TOD+OCF (n=4)
*
12-45 m
1(s)/0
IV
Frankel Grade (n=4) Lorenzo grade
All improved RD: improved BD: normal All recovered fully
13
III
Nurick’s grade (1-3)
3
IV
Muscle power grade 4 (n=4)
D
irreducible AAD: AD+ PF 18m (n=9), reducible AAD: PF (n=4) TAS/OCF 12 m
21.53±21.37 m
12 improved/1 worse
0/1(S)
2 y (2–4y )
0/0
CTT +brace (n=3)
43 m(16-65m)
10 improved 3 remained All improved
19 m
0/0
continue Author
nation
Age
Number of patients （F/M） ）
level
neurological signs before treatment
Treatment
Anti-TB treatment
Follow-time
neurological signs after Death/ treatment treatment failure
Gupta[12] 2006 Shukla[14] 2005
India
9-55y
11(4/7)
IV
>1y
1(C)/0
8-78y
16
IV
CTT +Halo jacket (n=11) TOD+ PF (n=10); PF (n=6)
18 m
India
muscle power Grade 0-4+(n=11), BD(n= 1) quadriparesis (n=16) BD(n=6)
Sinha[24] 2003 Behari[25] 2003
India
17-45y
IV
1(s)/0
10-60y
IV
Ranawat score IIIA (n=9), IIIB (n=9) Grade of neurological disabilities II(n= 8), III (n=1), IV(n= 9) RD (n=9) , BD (n= 4)
TRD + C1-2 wire
India
18 (8/10) 18
Arunkumar[6] 2002
India
17-55y
9(1/8)
IV
Nurick grade 3.4(range 1–5).
Krishnan[26] 2001
india
*
3
IV
neurologic deficit
Dhammi[27] 2001 Lal[2] 1992
India
20-45y
IV
india
23y
3 （2/1） ） 4(1/3)
IV
Muscle power grade 0-3(n=3), BD(n=1) Muscle power2-4
Lifeso[8] 1987 Fang [28] 1985
Audi Arabias china
14-65y
5
IV
Cord involvment
3-51y
4
IV
MRC grade I-IV
TOD+fusion: 3 PF:1
Tuli[29] 1974
india
*
8
IV
Neural complcations
CTT + brace
18 m
9.3m (1-36 ）
All recovered fully BD:* All improved BD: only 1 improved.
18 m
2 to 5 y
All recovered fully
0/0
C: Minerva jacket or 21 m hard collar (n =7). S: PF(n=7) /TOD + PF (n=4)
2.5y（6 m-7y） ） All improved. RD improvement. BD:*.
1(s)/1(s)
TOD + OCF (n=2), TORO + OCF(n=5), C1-2 decompression + OCF (n=2) CTT + bracing
18 m
18.8 m (7–46)
Nurick grade 2.3
0/0
*
2y
2 improved 1 remained
0/1(C)
22,48m
All complete recovery
0/1(C)
28m(3-48)
All complete recovery
1(s)/0
36m
Recovery
1(s)/0
15m
15m
recovery
0/0
*
1-7y
7 recover completely 1 partial recovery
0/0
CTT+ brace (n=3)
18 m
C: Minerva jacket 18m (n=1) S: drainage of RPA (n=1), PF+ minerva jacket (n=2) Surgery: OCF+Halo 15m
ASIA: Asia impairment Scale; TORO: Trans oral resection of odontoid; TOD: trans-oral decompression; TRD: transcervical retropharyngeal decompression; OCF: Occipitocervical fusion. RD: Respiratory distress; BD: Bladder disturbance; TAS: transarticular screws; *: Not reported; PF: posterior fixation. CTT: cervical tongs traction; ENAD: endonasal approach decompression;
ACRPD: anterior cervical and retropharyngeal debridement; OCFLS: occipital cervical fusion by laminar screw; OCFLSPS: occipital cervical fusion by laminar screw and pedicle screw
Table 2 the results of the conservation and the surgery. Method of treatment
No. Patients
Treatment Failure
Deaths
improving Muscle power
Complications
Conservation
94
14 (4 patient for non-improving neurological deficit, 10 patients for mobile AAD)
1
90
12 (pressure sores: 5, fixed atlantoaxial dislocation : 6 , sacral decubitus ulcer: 1)
surgery
153
2 (brooks technique with fracture of the C1 arch, neurological deterioration after ACDF)
5
145
12 (Anterior approach: wound infection: 2; Laryngeal nerve palsy: 2; transient dysphagia: 2; bone grafts displaced:2. Oral wound dehiscence: 1. Posterior approach: wire breakage: 1; fracture of the C1 arch: 1;；infection of the iliac bone graft donor site：1 )
Table 3 the results of the anterior decompression plus posterior fusion fixation and the posterior approach Procedure
No. Patients
Treatment Failure
Deaths
Non-Improving Muscle power
A+P P P value
89 62
1 1 P>0.05
3 2 P>0.05
3 5 P>0.05
A + P: anterior decompression plus posterior fusion fixation, P: posterior approach.
Records identified through database searching(n=281)： pubmed(n=130);EMBASE(n=151); Cochrane library(n=0)
Records after duplicates removed (n =141)
Records excluded according to title and abstract (n=86)
Full-text articles assessed for eligibility (n=57)
Studies included in quantitative analysis (n=24)
text articles excluded with reasons： No neurological defects =5 Insufficient data=4 Less than 2 patients=23 Overlapping cohort=1
Figure 1: Flow diagram showing the selection process
Abbreviations TB, tuberculosis; AAD, atlantoaxial dissociation; OR, odds ratio；PRISMA, Preferred Reporting Items for Systematic Review and Meta-analysis ; CI, confidence interval;