Classical Imaging Findings in Spinal Arachnoiditis

A 20-year-old female presented with progressively increasing low back pain of 3 months duration accompanied with nonspecific dysesthesias of both legs for 1 month. She developed weakness of both lower limbs with inability to walk 3 weeks prior to admission and also had urinary retention and constipation for 1 week for which she had been catheterized. Examination revealed flaccid paraparesis with power of Grade 1–2/5 in all groups of both lower limbs. Deep tendon reflexes were lost, and there was asymmetric sensory loss in all dermatomes below L1 in both legs. The plantars were bilaterally unresponsive. There was no spinal tenderness and no restriction or pain elicited on straight leg rising. No history of recent fever nor any past history of vertebral osteomyelitis, meningitis, or tuberculosis elsewhere in the body was present. A provisional diagnosis of a conus–epiconus tumor or a demyelinating polyradiculoneuropathy was thought of. However, a magnetic resonance imaging (MRI) study showed signal changes in the dorsal spinal cord with clumping of nerve roots and small cystic loculations in the lumbar spine on sagittal cuts. Axial imaging showed thickening of nerve roots with “empty thecal sac” sign. Contrast-enhanced imaging showed diffuse meningeal enhancement in the dorsal spine and soft tissue and root enhancement in the thecal sac in the lumbar spine [Figures 1 and 2]. A diagnosis of spinal arachnoiditis was then made.Figure 1: Sagittal magnetic resonance imaging showing (a) signal changes in the dorsal cord on T2 sequences with irregular notched cord outline (red arrows); (b) thickened clumped nerve roots in the lumbar spine on T2 sequences with multiple small caviations in the swollen conus (green arrow); (c) contrast enhanced images of dorsal spine showing diffuse meningeal thickening with intraconal enhancement (yellow arrow); and (d) contrast-enhanced images of the lumbar spine showing enhancement of nerve roots and soft tissue in the thecal spaceFigure 2: Axial magnetic resonance imaging T2 sequences showing (a) small cystic loculations around the terminal end of the cord (pink arrows); (b) grossly thickened conglomerated roots at L3 level in the spinal subarachnoid space; and (c) empty thecal sac at L5 level with traversing roots adherent to the dura at the sides (red arrows). Axial contrast-enhanced imaging (d) at L2 level shows diffuse enhancement of exudates with roots traversing through themA lumbar puncture was done that revealed a protein level of 602 mg% and sugar of 51 mg% (against a corresponding blood sugar of 138 mg%). Cell count was 64/mm3 of which 76% were lymphocytes. Aerobic cultures showed no growth of any microorganism, but tuberculosis polymerase chain reaction (PCR) GeneXpert was positive for Mycobacterium tuberculosis and showed no rifampicin resistance. Tubercular spinal arachnoiditis was diagnosed and she was put on standard antitubercular treatment and steroids. DISCUSSION Spinal arachnoiditis is known to mimic Guillain–Barre syndrome[1] or an intradural tumor,[2] which were the initial diagnoses we had considered. It can occur after surgery, trauma, meningeal carcinomatosis, and infections like tuberculosis. Tubercular spinal arachnoiditis develops secondary to (1) intracranial tubercular meningitis (drop down of infection from intracranial basal cisterns), (2) primary tubercular radiculomyelitis (blood borne), or rarely (3) tubercular vertebral spondylodiscitis (contiguous spread).[3] However, we could elicit no previous history of either meningitis or spinal vertebral tuberculosis in our case. Chang et al.[3] stated that unlike other causes of arachnoiditis, the spinal cord, meninges, and nerve roots are all frequently involved in tuberculosis as was seen in our case too. Ross et al.[4] have described three patterns of arachnoiditis on MRI in the lumbar spine: Type 1 where there are large conglomerations of thickened roots residing centrally in the thecal sac, Type 2 where the roots are stuck to the dura peripherally giving the appearance of an empty thecal sac, and Type 3 where there is intrathecal soft tissue. All these three classical patterns were seen in our patient in the lumbar spine. Spinal cord changes on MRI are most frequently found in the dorsal cord spanning several segments and include hyperintensity on T2-weighted images (either due to myelitis or vasculitis), swelling of and cavitation in the cord, variable enhancement of the cord on contrast imaging, irregular cord-cerebrospinal fluid (CSF) interface, and meningeal enhancement.[5] Enhancement of nerve roots on contrast and loculations in the CSF spaces has also been described. All of these were seen in our patient too. While routine CSF studies may point toward the possibility of tuberculosis, the diagnosis was historically confirmed by biopsy of intrathecal soft tissue. Currently, however, automated PCR is an authenticated method to detect mycobacterial DNA,[6] and we have used the same to clinch the diagnosis in our case as well. Previously, computed tomography myelography was used to diagnose spinal arachnoiditis. However, MRI with the classical findings mentioned above is now the gold standard as it enables evaluation of the whole spine irrespective of any spinal block, can pick up any associated spinal cord involvement,[5] and rules out other differential diagnoses. Declaration of patient consent The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient has given her consent for her images and other clinical information to be reported in the journal. The patient understands that her name and initials will not be published and due efforts will be made to conceal identity, but anonymity cannot be guaranteed. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.