Magnetic resonance imaging evaluation of posterior spinal tuberculosis: a cross-sectional study

Abstract

Background

Isolated posterior spinal (element) tuberculosis (TB) is uncommon compared to classical anterior spinal or para-discal TB. Here, we report magnetic resonance imaging (MRI) findings of posterior spinal TB in 19 patients without involvement of the vertebral body and intervertebral disc.

Purpose

To evaluate the MRI findings in isolated posterior spinal (element) TB.

Material and Methods

Clinical and MRI data of 19 patients of isolated posterior spinal TB were retrospectively evaluated.

Results

Of the 19 patients, group A comprised 4 (21%) patients with rapid onset lower limb weakness and pyramidal signs while group B comprised 15 (79%) patients without any neurological deficit. Lumbar vertebrae commonly affected 9 (47.4%) patients followed by dorsal vertebrae in 8 (42.1%) patients and cervical vertebrae in 2 (10.5%) patients. The pedicle was most commonly involved in 12 (63.2%) patients followed by the lamina in 11 (58%) patients, and spinous process and facet joint in 6 (31.6%) patients each. Extra-spinal inflammation/pyomyositis/paraspinal abscess was found in 13 (68.4%) patients followed by epidural abscess 3 (15.8%) patients and both extra spinal inflammation and epidural abscess in 3 (15.8%) patients (15.8%). Compressive cord myelopathy was observed in 4 (21%) patients, where three patients underwent emergency decompression laminectomy and the remaining 16 patients were treated conservatively with anti-tubercular therapy.

Conclusion

Initial diagnosis of isolated posterior element TB is challenging and requires a high index of suspicion. Early diagnosis of isolated posterior spinal TB is important as early treatment may be beneficial and decreases patient morbidity.

Keywords

Magnetic resonance imaging pain spinal cord

Introduction

Spinal tuberculosis (TB; Pott’s spine) comprises about 60% of all cases of bone and joint TB. In the classical form of tubercular spondylitis, the para-discal anterior body affection is seen (1); however, posterior extension of anterior classical tubercular spondylitis occurs in the posterior vertebral element in up to 24% of cases identified on magnetic resonance imaging (MRI) (2) and which are not categorized as posterior spinal TB. The exact incidence of isolated posterior element (spinal) TB is unknown. Isolated posterior element TB occurs in <1%–6% of spinal TB (1 –5). The incidence is < 2% in non-endemic areas and 5%–10% in endemic areas (4,6).

Posterior element TB commonly affects the thoracic vertebral arch followed by the thoracolumbar, lumbar, and cervical vertebral arch (7,8). Posterior element bony destruction or erosions are associated with epidural abscess and/or pyomyositis or abscess in posterior spinal muscles. Because of the difficulty in interpreting plain radiographic findings of posterior element TB or being occult, the diagnosis is usually delayed or missed (2 –4,7,9).

Posterior spinal TB is classified into primary and secondary types. The primary type only affects the posterior vertebral arch with a normal vertebral body and disc. This primary type is very rare (<1% of spinal TB). The secondary type is more common, where posterior vertebral body TB spreads into the posterior vertebral arch. Previous studies reported that involvement of the pedicle is the most common site (4,10), while some previous studies reported lamina involvement is most common (7,10).

Posterior spinal TB is usually associated with extradural granuloma and granulation tissue of which may encircle the dural sac causing secondary spinal canal stenosis (4,7). Thick granulation tissue around dural sac may produce a spinal tumor-like syndrome (11). Rarely, posterior spinal TB may present with anterior as well as posterior epidural abscesses with subsequent spinal cord compression with or without compressive cord myelopathy (12,13). Posterior spinal TB is associated with a high incidence of neurological deficit because of the intraspinal extension of tubercular granulation tissue or epidural abscess (4,14).

Posterior spinal TB is notorious for causing early spinal cord compression with early compressive effects over sensory posterior spinal horn, even though paraplegia is more common than sensory defect. Paraplegia is more common in classical anterior spinal TB (4,15).

Ultimately, this atypical posterior spinal TB results in spinal cord or cauda equina compression (7). Its unusual imaging features can result in a delay in its diagnosis and management (16). Therefore, early diagnosis of atypical posterior spinal TB can be sought with MRI, which helps in early surgical intervention in required patients, resulting in dramatic neurological recovery.

The aim of the present study was to evaluate the MRI appearances of posterior spinal TB.

Material and Methods

After approval from the institutional ethics review committee, a hospital-based retrospective study was conducted in a tertiary care hospital from December 2016 to December 2018 with a retrospective review of clinical and cross-sectional imaging data of 19 consecutive patients with posterior spinal TB without involvement of the vertebral body and intervertebral disc. MRI scans were performed in all patients and findings were correlated with histopathology of a computed tomography (CT)-guided biopsy sample or cytology/TB-polymerase chain reaction (PCR) test of a needle aspiration sample.

The MRI findings were categorized into three groups based on the site of involvement of the posterior vertebral element.

Patient selection

We included both outpatients and inpatients of both sexes presenting with lower backache, radiculopathy, paraparesis, paraplegia, urinary problems, and suspected cases of spinal TB. Informed consent was obtained from patients/guardian before undergoing MRI scan and CT-guided procedures.

Exclusion criteria

Patients with tubercular lesions affecting the craniovertebral junction and classical anterior spinal TB were excluded.

Detailed clinical findings were recorded and included demographic details, duration of symptoms, and presenting symptoms. The general physical and detailed neurological examination findings were recorded. According to Tuli (1), on the basis of neurological deficits, these patients were classified into four stages: stage I, in which the patient is unaware of the neurological deficit = neurological examination detects extensor planter reflexes and or ankle clonus; stage II, in which the patient is aware of the neurological deficit but is able to walk with support; stage III, in which the patient is not able to walk because of paralysis and sensory deficit that is < 50% normal; stage IV, in which the patient is not able to walk and has flexor muscle spasm or paralysis in flexion, flaccid paralysis, sensory deficit that is > 50% normal, or bladder and bowel involvement.

MRI protocols

Nineteen patients were subjected to an MRI examination using Siemens Avanto 1.5-T B15 machine (Siemens Medical Systems, Erlangen, Germany).

MRI of the lumbosacral spine was performed in the sagittal, coronal, and axial planes using a combination of pulse sequences. MRI was performed with the patient in the supine position.

A combination of transverse, sagittal, and coronal images of the lumbosacral spine were obtained using the following sequences: (i) sagittal T1-weighted (T1W) spin-echo sequence (TR/TE = 450–500/9–15 ms, matrix size = 512 × 512) without fat saturation and slice thickness of 3–4 mm; (ii) sagittal T2-weighted turbo spin-echo sequence (TR/TE = 3400–4600/110–120 ms, matrix size = 512 × 512) with slice thickness of 3–4 mm; (iii) sagittal and coronal short tau inversion recovery (STIR) sequence (TR/TE = 4500–4900/25–28 ms, TI = 160) with slice thickness of 3–4 mm; (iv) sagittal diffusion-weighted imaging (DWI) sequence (TR/TE = 2500–3000/86–100 ms, slice thickness =3 mm, b value = 0 and 1000 s/mm², interslice gap =3.3 mm, matrix = 128 × 128, flip angle = 90°) – DWI was always obtained before contrast administration and was acquired by a single shot spin-echo EPI sequence; (v) sagittal gradient recalled echo (GRE) sequence: T2_me2d sequence (TR/TE = 550–600/20–26 ms, matrix size = 256 × 256) with slice thickness of 3–4 mm; (vi) sagittal three-dimensional (3D) T2-SPACE (sampling perfection with application optimized contrasts by using flip angle evolution) sequence (TR/TE = 900–1000/80–130 ms, flip angle = 140°–150°, matrix size = 256 × 256) with slice thickness of 1 mm; and (vii) multiplanar post-contrast T1W spin echo sequence with fat suppression was acquired after injecting i.v. gadopentetate dimeglumine at a dose of 0.1 mmol/kg body weight (TR/TE = 3400–4600/110–120 ms, matrix = 512 × 512) with slice thickness of 3–4 mm.

MRI evaluation

Nineteen patients with isolated posterior spinal (element) TB without involvement of vertebral body and intervertebral disc were examined. The evaluation was carried out to look for the level of vertebral involvement, site of the posterior element affection, bone erosion/marrow changes in the posterior elements, associated pyomyositis in the paraspinal muscles, epidural abscess, and spinal cord compression. Patterns of post-contrast enhancement in affected posterior vertebral elements and associated surrounding inflammatory components were also sought.

CT-guided procedure

CT-guided core needle biopsy samples were obtained by using either an 11-gauge or 13-gauge Osteo-Site® bone biopsy needle (Cook Medical, Bloomington, IN, USA). CT-guided aspiration procedures were done using an 18–20-gauge lumbar puncture needle.

Laboratory test

All patients were evaluated with complete blood count (CBC) with erythrocyte sedimentation rate (ESR) and ELISA for HIV. A posteroanterior (PA) view of chest X-ray was done in all patients.

Five patients underwent CT-guided core biopsy where samples were evaluated for histopathological examination. Eleven patients underwent CT-guided needle aspiration and three patients underwent emergency decompression laminectomy. The aspirated samples or postoperative specimen were evaluated for histopathology, acid-fast bacilli stain, culture, or PCR for Mycobacterium tuberculosis.

Treatment

The diagnosis of isolated posterior spinal TB was established by clinical, MRI findings and pathological confirmation from guided aspiration, CT-guided biopsy sample, or sample obtained after decompression laminectomy. The patients were divided into two groups. Group A comprised 4 (21%) patients with rapid onset lower limb weakness, pyramidal signs, and MRI findings of posterior vertebral element affection with epidural abscess/intraspinal inflammatory component. Of the four patients, three underwent emergency decompressive laminectomy followed by anti-tubercular therapy (ATT) while the other patient was treated conservatively with ATT. Group B comprised 15 (79%) patients without any neurological deficit, had pyomyositis of posterior paraspinal muscles with posterior element affection without epidural abscess/intraspinal inflammatory component. These 15 patients were treated conservatively with ATT only.

All patients were treated with an ATT four-drug regimen (Rifampicin 600 mg, Isoniazid 300 mg, Ethambutol 1000 mg, and Pyrazinamide 1500 mg) for the first three months followed by three drugs, i.e. Rifampicin, Isoniazid, and Ethambutol, for nine months. The clinical, hematological, and radiological parameters showed significant regression of the lesions on subsequent follow-ups at three, six, nine, and twelve months.

Statistical analysis

Data were presented in terms of percentage and mean. Calculations were done using Microsoft Excel and SPSS programs (Statistical Package for the Social Science version 16, SPSS Inc., Chicago, IL, USA).

Results

The study sample comprised 19 patients (mean age =32.5 ± 1.5 years; male:female ratio = 2.8:1). Various MRI findings of isolated posterior spinous TB in 19 patients of were summarized in Tables 1 and 2. Affection of the C4 to L5 vertebral level was observed in this study sample. The majority of patients presented with backache and spastic lower limb weakness. The patients’ clinical presentations were categorized according Tuli staging: 5 (26.3%) patients presented with stage I; 10 (52.6%) patients presented with stage II; 1 (5.3%) patient presented with stage III; and 3 (15.8%) patients presented with stage IV. Group A comprised 4 (21%) patients with rapid onset lower limb weakness, pyramidal signs, and MRI findings of posterior vertebral element affection with epidural abscess causing compressive cord myelopathy, where three patients underwent emergency decompressive laminectomy followed by ATT and one patient was treated conservatively with ATT. Group B comprised 15 (79%) without any neurological deficit, with pyomyositis of the posterior paraspinal muscles with posterior element affection, and without compressive cord myelopathy and were treated conservatively with only ATT.

Table 1.

MRI findings in 19 patients with posterior element TB.

Patient no.	Age (years)/Sex	Tuli staging clinical features	MRI findings	Affected vertebra	Guided procedure	Management
1	45/M	II	Bilateral lamina, spinous process, right articular facet involvement with bilateral para-spinal inflammatory components	L5	CT-guided biopsy	Only ATT
2	11/M	II	Right pedicle, right paravertebral abscess, posterior epidural abscess in lower lumbar area	D12	CT-guided aspiration	Only ATT
3	45/M	II	Left pedicle, left transverse process, left spinous process, left paravertebral abscess, edematous left psoas muscle	L3	CT-guided biopsy	Only ATT
4	36/F	II	Spinous processes with posterior paraspinal abscesses	L2, L3, and L4	CT-guided aspiration	Only ATT
5	25/M	III	Spinous process, left-sided C6/7 facet joint with bilateral paraspinal inflammatory and posterior epidural abscess extending into dorsal region causing cord myelopathy	C6	CT-guided aspiration	Only ATT
6	16/M	I	Right-sided pedicle of L4 and L5 vertebrae, right L5 transverse process and right L4/5 facet joint with nearby right paraspinal abscesses	L4 and L5	CT-guided aspiration	Only ATT
7	22/ M	II	Right pedicle and right paravertebral inflammation with post-biopsy tract	D12	CT-guided biopsy	Only ATT
8	58/M	II	Right pedicle, lamina and right L3/4 facet joint, right paraspinal and psoas abscesses, anterior and posterior epidural abscesses	L4	CT-guided aspiration	Only ATT
9	50/M	I	Right lamina with nearby right paravertebral inflammatory components	L5	CT-guided aspiration	Only ATT
10	16/M	II	Left-sided pedicles, laminae, transverse processes and left sided C4/5 and C5/6 facet joints, inflammatory changes seen in left paravertebral region and left paraspinal muscles	C4, C5, and C6	CT-guided aspiration	Only ATT
11	15/F	II	Spinous process of L2 vertebra with surrounding inflammatory paraspinal muscle inflammatory components	L2	CT-guided aspiration	Only ATT
12	33/F	II	Bilateral transverse processes, laminae, and right-sided pedicle with paraspinal inflammatory components	D7 and D8	CT-guided aspiration	Only ATT
13	50/F	IV	Bilateral laminae and spinous process of D3 vertebra with nearby inflammatory collections, posterior epidural abscess causing dorsal cord compressive myelopathy	D3	-	Decompressive laminectomy
14	34/M	I	Right pedicle of L4 and L5, left lamina, and articular facet of L3 vertebra	L3, L4, and L5	CT-guided biopsy	Only ATT
15	17/M	II	Right transverse process of L3 with right posterior paraspinal abscess formation	L3	CT-guided aspiration	Only ATT
16	59/M	I	Right pedicle of L3 vertebra with abscesses in right paravertebral region	L3	CT-guided aspiration	Only ATT
17	19/F	IV	Right pedicle and lamina of D3 with posterior epidural abscess formation and compressive cord myelopathy	D3	–	Decompressivelaminectomy
18	33/M	IV	Right pedicles of D6 and D7 vertebrae with a long segmental posterior epidural space abscess in dorsal level with T2 hyperintense cord compressive myelopathy	D6 and D7	–	Decompressivelaminectomy
19	36/M	I	Left pedicle and posterior lateral body of D12 vertebra with affection of left costovertebral articulation and posterior end of left 12th rib	D12	CT-guided biopsy	Only ATT

ATT, anti-tubercular therapy; CT, computed tomography; MRI, magnetic resonance imaging; TB, tuberculosis.

Table 2.

Details of posterior vertebral element involvement on MRI in 19 patients of isolated posterior element TB.

Site of involvement of posterior vertebral element	Patients (n = 19)
Lamina	11 (58)
Pedicle	12 (63.2)
Spinous process	6 (31.6)
Transverse process	5 (26.3)
Facet articular process	6 (31.6)
Cervical vertebral arch	2 (10.5)
Dorsal vertebral arch	8 (42.1)
Lumbar vertebral arch	9 (47.4)
Spinal cord compression	6 (31.6)
Spinal cord myelopathy	4 (21)
Epidural abscess	6 (31.6)
Paraspinal pyomyositis/abscess	16 (84.2)

Values are given as n (%).

The posterior vertebral element involvement was categorized into three groups. Group 1 comprised 3 (15.8%) patients with only affected posterior arch processes such as spinous process (Fig. 1), transverse process, and apophyseal (facet) joint process. Group 2 comprised 7 (36.8%) patients with affected posterior arch such as the lamina and pedicle (Fig. 2) and Group 3 comprised 9 (47.4%) patients with mixed involvement of posterior arch process and posterior arch (Fig. 3).

Fig. 1.

A 36-year-old female patient with backache and back swelling. (a) Sagittal STIR image shows hyperintensities in the spinous processes of L2, L3, and L4 vertebrae with surrounding hyperintense collection and overlying subcutaneous tissue edema (arrow). (b) Sagittal T1W image shows hypointensities in the affected spinous processes. (c–e) Post-gadolinium sagittal, and axial T1W images showed homogenous enhancement in the affected spinous processes with peripherally enhancing collections around the spinous processes along with phlegmonous enhancing inflammatory components in bilateral paraspinal regions and peripherally enhancing collection over the tip of spinous process (arrow). T1W, T1-weighted.

Fig. 2.

A 36-year-old man presented with backache. (a) Coronal STIR image showed hyperintensities in the left pedicle and posterior lateral body of D12 vertebra (arrow). (b, c) Post-gadolinium coronal and sagittal images showed irregular patchy post-contrast enhancement in the affected regions of vertebra with minimally enhancing inflammatory component in the left paravertebral region (arrow). (d, e) Axial post-gadolinium images showed the irregular enhancement in the affected vertebra, pedicle, and left-sided costovertebral articulation and posterior end of the left 12th rib (arrow) with surrounding inflammatory changes.

Fig. 3.

A 45-year-old male patient presented with backache for five months. (a) Sagittal STIR image showed ill-defined STIR hyperintensities in the posterior element of L5 vertebra (arrow). (b, c) Axial T2W and T1W images showed T2 hyperintensities and T1 hypointensities in the bilateral lamina, right articular facet, and spinous process (arrow). (d, e) Axial post-gadolinium T1W images showed ill-defined enhancement in the affected posterior vertebral elements with enhancing phlegmonous inflammatory component in bilateral paraspinal regions, more on the right side (arrow). T1W, T1-weighted; T2, T2-weighted.

Isolated posterior spinal TB was found in the lumbar vertebra (Fig. 2) in 9 (47.4%) patients followed by the dorsal vertebra (Fig. 4) in 8 (42.1%) patients and the cervical vertebrae in 2 (10.5%) patients (Table 2). The isolated posterior vertebral element affecting a single vertebra was observed in 13 (68.4%) patients (Figs. 2 and 4) followed by two vertebra in 3 (15.8%) patients and three vertebrae (Fig. 1) in another 3 (15.8%) patients. Pedicle involvement was observed in 12 (63.2%) patients, unilaterally in 11 (57.9%) patients (Figs. 2 and 4) and bilaterally in 1 (5.3%) patient. Involvement of the lamina was seen in 11 (58%) patients, unilaterally in 8 (42.1%) patients (Fig. 3) and bilaterally in 3 (15.8%) patients. Spinous process (Fig. 1) and involvement of the facet joint (Figs. 3 and 5) were seen in 6 (31.6%) patients each. Transverse process involvement (Fig. 5) was observed in 5 (26.3%) patients.

Fig. 4.

A 22-year-old man presented with backache for five months. (a) Sagittal STIR image showed ill-defined hyperintensities in the right pedicle of D12 vertebra (arrow). (b–d) Sagittal and axial post-gadolinium images showed moderate enhancement of the right pedicular lesion (arrow) with enhancing inflammatory component around the hypointense post biopsy tract in the right paraspinal region (arrow).

Fig. 5.

A 16-year-old male patient with backache. (a) Sagittal T2W and (b, c) sagittal and coronal STIR images showed ill-defined mixed signal intensities in posterior vertebral elements of L4 and L5 vertebrae with affection of right L4/5 facet joint with surrounding T2 hyperintense collections (arrow). Post-gadolinium (d) coronal and (e, f) axial images showed patchy enhancement in right-sided L5 transverse process (arrow) with phlegmonous enhancing component around the affected transverse process and peripherally enhancing abscess in right paraspinal region (arrow).

Abnormal T2 and STIR hyperintense signal intensities with various degrees of bony destructions/erosions was observed in all patients with isolated posterior spinal TB. Extra-spinal phlegmonous inflammatory component/pyomyositis/abscess was found in 13 (68.4%) patients (Fig. 5), only epidural abscess in 3 (15.8%) patients, and both extra spinal inflammatory components and epidural abscess in 3 (15.8%) patients (Fig. 6). Epidural abscess displaced spinal cord displacement without T2 hyperintense compressive myelopathy in 2 (10.5%) patients, with T2W hyperintense compressive cord myelopathy in 4 (21.1%) patients. T2W hyperintense spinal cord compressive myelopathy was observed in 2 (10.5%) patients in group 3 and 2 (10.5%) patients in group 2 with isolated posterior spinal TB. However, no statistical significance was observed within or between the groups causing spinal cord compressive myelopathy (P = 0.76). A PA view of the chest X-rays showed concurrent pulmonary TB in 5 (26.3%) patients. For isolated posterior spinal TB, 3 (15.8%) patients were treated with emergency decompression laminectomy followed by ATT for a duration of 12 months and the remaining 16 (84.2%) patients were treated conservatively with only ATT for 12 months. Patients in group A had poorer outcomes than those in group B. Two patients in group A continued to have spastic paraplegia in the follow-up period while the other two patients in group A showed a significant improvement in neurological deficit. All patients in group B remained neurologically intact during the follow-up period.

Fig. 6.

A 58-year-old male patient with backache and paraparesis. (a, b) Sagittal and STIR images showed ill-defined hyperintensities in posterior vertebral elements of L4 vertebra (arrow). (c) Sagittal DWI and (d) ADC map images show diffusion restriction (arrow) in paraspinal lesion with low ADC value. (e, f) Sagittal post-gadolinium images showed peripherally anterior and posterior (arrow) epidural abscesses along with phlegmonous enhancing inflammatory component in right paraspinal region with abscess formation (arrow). (g, h) Axial post-gadolinium images show patchy enhancement in the right pedicle and right lamina with epidural, right paraspinal and right psoas abscesses. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging.

Discussion

Pott’s spine is a common site for extrapulmonary TB. Three distinct patterns of common spinal TB occur sicj as para-discal, central, and anterior in decreasing frequency (6,17). However, posterior spinal (element) TB is rare and has atypical features (3,6,7,10,14). The increasing frequency of atypical TB, including isolated posterior spinal TB, is reported in people with HIV, drug addicts, or immunocompromised patients (18,19,21). Narlawar et al. (2) found 7/21 (33.3%) patients who were positive for HIV had isolated TB of the posterior elements.

Plain radiography and CT scan play an important role in characterizing the typical form of spinal TB (14,17). However, posterior spinous TB is usually diagnosed in the later stages where more destruction of posterior vertebral elements occurs (14,17,21). MRI is the imaging of choice for detailed evaluation of posterior spinal destruction, identification of granuloma or granulation tissues, paravertebral inflammation/abscess, epidural abscesses, and spinal cord compression (22). Sometimes the clinical features of posterior spinal TB resemble extradural malignancy, and cross-sectional imaging, especially MRI, can differentiate the two (23).

MRI can assist in the early detection of STIR hyperintense bone marrow edema in the affected posterior vertebral element, adjacent paraspinal soft-tissue edema, extradural granulation tissue, intraspinal inflammation/abscess, and spinal cord compression. However, CT can detect the bony destruction and surrounding inflammatory changes except for the spinal cord compressive myelopathy. Hence MRI must be utilized in suspected cases of posterior spinal TB or cases with an uncertain diagnosis to avoid delayed diagnosis and poor patient outcome.

In spinal TB, the infective foci in the vertebra result in the initial destruction of osteonecrosis of the cancellous bone from septic micro-emboli followed by destruction of the cortical bone and subsequent involvement of adjacent vertebrae and intervertebral disc. Spread of infection extends beneath the longitudinal ligaments, causing the subsequent collapse of the affected vertebra and may be associated with prevertebral, paravertebral, and epidural abscesses (24).

Because of the rarity and unfamiliarity of the non-specific clinical and radiological features, the diagnosis of isolated TB of the posterior elements is commonly delayed (25). As with a delay in the diagnosis of isolated posterior element TB, subsequent affection of anterior vertebral elements occurs and leads to more severe spinal instability. Because of its rarity and the delay in diagnosis, isolated posterior element TB can sometimes be misdiagnosed as a neoplastic lesion (9). The delay and error in the diagnosis of isolated posterior element TB accounts for the continuing morbidity of an otherwise curable disease.

Therefore, increased awareness and imaging features of the atypical features of spinal TB and atypical isolated posterior element TB lead to improved diagnostic accuracy in the differential diagnosis of primary or metastatic spinal tumors at an early stage of the disease process, i.e. before the occurrence of irreversible neurologic deficits and spinal deformities (9).

Naim-ur-Rahman (3) detected isolated posterior element TB in five patients over 10 years of study, where involvement of the lamina was the most common. Kumar (4) reported 21 patients with posterior element TB in 1985, where the pedicle was most commonly affected. Isolated involvement of the transverse process known as spondylitis tuberculosa lateralis was reported by Vincurova et al. (26).

However, in our study sample of 19 patients, involvement of the pedicle was more common (63.2%) followed by the lamina (58%). Spinal cord compression and compressive cord myelopathy was encountered in groups 2 and 3 in our study sample because of the early disruption of the bony spinal rim in these groups and where three patients underwent emergency decompression laminectomy followed by ATT for 12 months.

In conclusion, isolated involvement of the posterior elements is rare in a case of spinal TB. MRI is extremely useful in evaluating the extent of posterior vertebral element involvement and response to treatment. Therefore, early diagnosis of atypical posterior spinal TB with medical management with ATT and/or prompt surgical decompression helps in reducing patient morbidity or neurological deficit from spinal cord compression. Hence, isolated posterior spinous TB should be considered in the differential diagnosis of marked or massive destruction of posterior vertebral elements and when only a ghost of the posterior elements remains.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Deb K Boruah

References

Tuli

SM.

Tuberculosis of the spine In: Tuli

, editor. Tuberculosis of the skeletal system (bones, joints, spine and bursal sheaths). 3rd edn. New Delhi: Jaypee Brothers Medical Publishers (P) Ltd., 2004:191–343.

Narlawar

Shah

Pimple

, et al. Isolated tuberculosis of posterior elements of spine: magnetic resonance imaging findings in 33 patients. Spine (Phila Pa 1976) 2002; 27:275–281.

Naim-ur-Rahman

Atypical forms of spinal tuberculosis. J Bone Joint Surg Br 1980; 62:162–165.

Kumar

Clinical study and classification of posterior spinal tuberculosis. Int Orthop 1985; 9:147–152.

Arora

Sabat

Maini

, et al. Isolated involvement of the posterior elements in spinal tuberculosis: a review of twenty-four cases. J Bone Joint Surg Am 2012; 94:151.

Abdelwahab

Camins

Hermann

, et al. Vertebral arch or posterior spinal tuberculosis. Skeletal Radiol 1997; 26:737–740.

Babhulkar

Tayade

Babhulkar

SK.

Atypical spinal tuberculosis. J Bone Joint Surg 1984; 88:23–26.

Sharif

Aideyan

Clark

, et al. Brucellar and tuberculous spondylitis: comparative imaging features. Radiology 1989; 171:419–425.

Naim-ur-

Rahman

El-Bakry

Jamjoom

, et al. Atypical forms of spinal tuberculosis: case report and review of the literature. Surg Neurol 1999; 51:602–607.

10.

Bell

Cockshott

WP.

Tuberculosis of the vertebral pedicles. Radiology 1971; 99:43–48.

11.

Jain

Shanmugam

Atypical spinal tuberculosis-extra-osseous extra dural tuberculoma: A retrospective study. Int J Recent Trends Sci Tech 2015; 14:332–334.

12.

Kak

Pani

Chopra

JS.

Epidural spinal Tuberculoma presenting as “spinal tumour syndrome”. Tubercle 1972; 53:139–142.

13.

Griffiths

DLI

Seddon

Roaf

Pott’s paraplegia. London: Oxford University Press, 1956.

14.

Solomon

Sacks

Goldsmidt

RP.

Neural arch tuberculosis: A morbid disease-radiographic and computerized tomographic findings. Int Orthop (SICOT) 1995; 19:110–115.

15.

Monaghan

Gupta

Barrington

NA.

Case report: tuberculosis of the spine — an unusual presentation. Clin Radiol 1991; 43:360–362.

16.

Acharya

Ratra

GS.

Posterior spinal cord compression: outcome and results. Spine 2006; 31:574–578.

17.

Kapoor

Garg

Dhaon

, et al. Tuberculosis of the posterior vertebral elements: a rare cause of compression of the cauda equina. J Bone Joint Surg 2005; 87:391–394.

18.

Mallolas

Gatell

Rovira

, et al. Vertebral arch tuberculosis in two human immunodeficiency virus-seropositive heroin addicts. Arch Intern Med 1988; 148:1125–1127.

19.

Shafer

Edlin

BR.

Tuberculosis in patients infected with human immunodeficiency virus: perspective on the past decade. Clin Infect Dis 1996; 22:683–704.

20.

Mallolas

Gatell

Rovira

, et al

Vertebral arch tuberculosis in two human immunodeficiency virus-seropositive heroin addicts. Arch Intern Med 1988; 148:1125–1127.

21.

Ragland

Abdelwahab

Braffman

, et al. Posterior spinal tuberculosis: a case report. AJNR Am J Neuroradiol 1990; 11:612–613.

22.

Currie

Galea-soler

Barron

, et al. MRI characteristics of tuberculous spondylitis. Clin Radiol 2011; 66:778–787.

23.

Acharya

Ratra

GS.

Posterior spinal cord compression: outcome and results. Spine 2006; 31:574–578.

24.

Mulleman

Mammou

Griffoul

, et al

Characteristics of patients with spinal tuberculosis in a French teaching hospital. Joint Bone Spine 2006; 73:424–427.

25.

Yazdi

Rahimizadeh

An illustrative case Isolated posterior spinal arch tuberculosis. World Spinal Column J 2012; 3:98–101.

26.

Vincurova

Maar

Schwartz

[Mycobacterial infection at the transverse process of the L2 lumbar vertebra caused by Mycobacterium fortuitum]. Z Orthop ihre Grenzgeb 1984; 122:278–280.