Severity of foraminal lumbar stenosis and the relation to clinical symptoms and response to periradicular infiltration-introduction of the "melting sign".

BACKGROUND CONTEXT: Nerve root compression causing symptomatic radiculopathy can occur within the intervertebral foramen. Sagittal magnetic resonance imaging (MRI) sequences are reliable in detection of nerve root contact to intraforaminal disc material, but a clinically relevant classification of degree of contact is lacking.

PURPOSE: This study aimed to investigate a potential relation of amount of contact between intraforaminal disc material and nerve root to clinical findings and response after periradicular corticosteroid infiltration.

STUDY DESIGN: A post hoc analysis of a prospective cohort was carried out.

PATIENT SAMPLE: Patients who underwent computed tomography (CT)-guided periradicular corticosteroid infiltration (L1-L5) at our institution (January 2014 to May 2016) were included.

OUTCOME MEASURES: The medical records and radiographic imaging were reviewed.

METHODS: T2-weighted MRI of the lumbar spine of patients with single-level symptomatic radiculopathy with (responders, n=28) or without (non-responders, n=14) pain relief after periradicular infiltration with corticosteroids were measured and compared by two independent readers to determine the amount of intraforaminal nerve root contact with the intervertebral disc ("melting" of the T2-hypointense signal). Pain relief was defined with a pain level decrease of >50% on a visual analogue scale and lack of pain relief with a pain level decrease of <25%, respectively. The amount of T2-hypointensity melting of disc and nerve root was categorized to 0%, 1%-25%, and over 25%.

RESULTS: Reader one identified 0% T2-melting in none of the responders, 1%-25% melting in 13 patients (46.4%), 26%-50% in 15 of the 28 patients (53.6%) with pain relief after periradicular corticosteroid infiltration (responders), with a mean amount of T2-melting of 5.9±2.1 mm, whereas the non-responder group had 0% T2-melting in 2 patients (14.3%), 1%-25% T2-melting in 11 patients (78.6%), and 26%-50% in 1 patient (7.1%), with a mean amount of T2-melting of 2.6±1.9 mm (p<.05). Reader two identified 0% T2-melting in none, 1%-25% T2-melting in 15 (53.6%) patients, and 26%-50% in 13 of the 28 responders (46.4%), with mean amount of 6.3±1.9 mm. In the non-responder group 0% T2-melting was seen in 3 patients (21.4%), 1%-25% T2-melting in 10 patients (71.4%), and 26%-50% in 1 patient (7.1%), with a mean amount of T2-melting of 2.7±1.9 mm (p<.05). None of the MRI showed T2-melting in over 50% of the circumference of the intraforaminal nerve root. A T2-melting of >25% had a high specificity of 93% but a sensitivity of 50%, thus a positive likelihood ratio of 7.5, to identify those with a pain relief of more than 50% after infiltration.

CONCLUSION: The amount of T2-melting of disc material and nerve root on sagittal MRI (>25%) predicts the amount of pain relief by periradicular infiltration in patients with intraforaminal nerve root irritation.