The present prospective randomised study includes 40 consecutive patients, 20 men and 20 women, presenting with cervical radiculopathy. The mean age was 51 years (range 27–65), with mean symptom duration of 31 months (range 3–120) (Table ). The inclusion criteria were one-sided cervical radiculopathy with radicular distribution of arm pain distal to the elbow and corresponding significant degenerative pathology of the cervical spine, at one or two levels on the same side as the radicular pain and visualised by MRI. Only MRI pathology with a close relation to the nerve root(s) is classified as significant. A positive response to a diagnostic transforaminal SNRB, at level(s) presenting with MRI pathology on the same side as the radicular pain, was mandatory for all patients. In all 40 patients, the performed investigations indicated affection of cervical nerve roots based on a degenerative disease in the cervical spine. Patients with spinal cord compression and/or myelopathy were excluded. Informed consent was used. The ethics committee at the University Hospital of Lund, Sweden approved the study.
Patient data, randomised treatment, and treatment results
Patients referred to our department with cervical radiculopathy routinely undergo clinical examination by a neurosurgeon, with MRI investigation of the cervical spine. Patients may also undergo investigations with diagnostic transforaminal SNRB with 0.5 ml Carbocain® (Mepivacaine, 10 mg/ml, Astra, Sweden) at levels associated with significant degenerative pathology on MRI. Consenting patients having undergone these investigations that fulfilled the inclusion criteria were consecutively randomised for treatment with transforaminal injection(s) of the nerve root(s).
Prior to randomised treatment, MRI evaluation was assessed as follows: firstly, by a specially trained radiologist at the referring hospital. Secondly, a neuroradiologist at the Department of Neuroradiology at our hospital evaluated and demonstrated the MRI at a clinical conference with the neurosurgeons. Thirdly, the neuroradiologist performing the diagnostic SNRB reviewed the MRI prior to the root block procedure. Before the randomised treatment, the neuroradiologist performing the injections once again evaluated the MRI.
Selection of nerve roots for randomised treatment
Nerve roots on the same side as the radicular pain, at levels presenting with significant degenerative pathology on MRI and responding with a significant pain reduction to diagnostic SNRB, were randomised for treatment. Consequently, patients with positive response to SNRB at one level received the random treatment at one level, and patients with positive response to SNRB at two levels received the random treatment at both levels.
The 40 patients were randomised into two equal treatment groups. One group received treatment with 0.5 ml Carbocain and 1 ml Depo Medrol® (40 mg methylprednisolone acetate) per injection, and the control group received treatment with 0.5 ml Carbocain (Mepivacaine) and 1 ml saline per injection. Only the neuroradiologist performing the blocks was aware of the content of the injections. The patient and the evaluating neurosurgeon/physiotherapist were blinded.
Criteria of positive response to the diagnostic SNRB
If the patient reported a significant subjective arm pain relief (radicular pain) and presented at least 50% arm pain reduction on an visual analogue scale at follow up 30 min after the SNRB, the diagnostic SNRB was classified as positive [2
Performed by a neuroradiologist using fluoroscopy, the same transforaminal injection technique was used both for diagnostic block as well as for the randomised block [2
]. To determine the correct position for injections, thus minimising the risks of intravascular or intradural injection, a small amount of contrast medium was introduced prior to injection [3
Patients were evaluated immediately prior to and 3 weeks following the randomised injections. All evaluations were performed by a neurosurgeon and a physiotherapist, and included patient examination and completion of subjective symptom/reaction questionnaire. In three of the patients, the follow up had to be made by telephone.
A questionnaire including ten questions was used as follow up of the randomised treatment (Table ). The aim of the questionnaire was to cover subjective changes in symptoms associated with cervical radiculopathy, including location of pain, muscle strength, sensory changes, analgesic dose, and sleep quality. The patients were encouraged to report any additional effect, if present. Following the completion of the questionnaire, a clinical investigation was undertaken.
Design of follow-up questionnaire used to detect effect from the random treatment
In order to assess early as well as late changes from the randomised treatment, the patients were asked to relate changes of symptoms to the first, second, and third weeks.
The clinical criteria of positive response to the randomised treatment were a significant subjective reduction of the radicular pain and/or significant subjective reduction of neurological symptoms. When performing the statistical analysis of treatment result, all data from the questionnaire were compared (Table ).
Effect from the random treatment for all patients at follow up after 3 weeks
Questionnaire results were compiled according to first, second, and third weeks following treatment. All data were analysed using a computer software package (SPSS, release 11.0.0; SPSS, Chicago, IL, USA).