Price et al recommend that caudal epidural injections should be carried out withx ray screening.1 In practice the technique is often performed in the outpatient clinic, where screening facilities are not usually available.

I have reviewed the outcome of 44 unscreened caudal epidural injections given to 34 patients in my outpatient clinic in the 12 months from July 1999 to June 2000. Patient selection was principally on the basis of unilateral nerve root pain in the leg in the presence of symmetrical straight leg raising. However, nine patients had asymmetrically reduced straight leg raising and four of those with full straight leg raising had pain in the L4 dermatome, mostly with a positive femoral nerve stretch test on the affected side. Spinal mobility and the presence or absence of a neurological deficit did not influence the decision to offer an epidural injection.

The procedure was carried out with the patient lying prone on the examination couch. After thorough skin preparation and a local injection of 2 ml lidocaine (lignocaine) 2%, 40 mg triamcinolone hexacetonide mixed with 10 ml lidocaine 0.5% and 10 ml normal saline was injected into the epidural space through the sacral hiatus with a 20G spinal needle. The patient was subsequently kept prone for 30 minutes and remained horizontal for a total of 60 minutes before being allowed home.

Six of the patients were subsequently given a second epidural injection and two of them received three in the 12 month period. The procedure was repeated only if there was a satisfactory response to the original injection.

For the purpose of this audit, outcomes were graded as:

1 Complete or substantial relief of root symptoms for a minimum of three months.

2 Temporary relief for between one week and three months.

3 A sustained partial response (up to 50% reduction in pain).

4 No response (negligible pain relief or pain reduced for less than one week).

After the first epidural injection, 13 patients were cured of their root pain or obtained relief for three months or more, six derived temporary relief, four partial relief, and 10 showed no response. One of the non-responders was subsequently deemed to have Achilles tendinitis rather than S1 root pain. The outcome in the remaining case is unknown as the patient defaulted from follow up.

Of the eight patients who received a second epidural injection, three were cured or had sustained relief, three partial relief, and two did not respond (despite previous satisfactory responses for four and nine months, respectively). Of the two patients who were given a third epidural injection, one enjoyed sustained relief of pain while the other had temporary relief for six to eight weeks.

In brief, 23/34 (68%) of patients experienced at least a temporary or partial response to the initial unscreened caudal epidural injection and of the eight patients who were given either two or three epidural injections, four obtained sustained relief from their leg pain.

There were no complications from the procedure.

On further analysis of the results, 68% of patients with full straight leg raising showed at least a temporary response to the first epidural injection compared with 44% of those with asymmetrically reduced straight leg raising. Of patients with a neurological deficit, 85% experienced at least temporary pain relief.

Sixteen patients (47%) were ultimately referred to the chronic pain management services for nerve root blocks and three (9%) for a surgical opinion on the basis of the results of a magnetic resonance imaging scan of their lumbar spine.

Although x ray guidance theoretically might improve the accuracy of placement of the spinal needle, these results show that caudal epidural injections can be carried out safely and effectively in an outpatient setting without radiological screening. This also ensures prompter treatment for the patient.