A reply

injection of contrast medium. Satisfactory placement was achieved on each occasion simply by withdrawal of the needle by 1 or 2 mm, and the infiltration was performed uneventfully . The anterior facet joint capsule is formed by the ligamentum flavum,’ so it is very probable that overpenetration of the joint will result in intrathecal injection. We strongly support the recommendations of Goldstone and Pennant about the availability of resuscitation equipment. In addition, we suggest that bony contact with the most posterior part of one of the articular facets should be deliberately sought during needle placement, to act as a ‘depth gauge’ for subsequent needle advancement which should be no more than 2 or 3 mm beyond this level. It is interesting that the authors’ use of low-volume facet arthrography did not prevent this complication. We have found the images obtained by facet arthrography to be sometimes difficult or impossible to interpret with confidence, even when volumes of 0.5-1 ml contrast medium are employed. Frequent aspiration is prudent even after unequivocal arthrography, since most patients tend to shift position during this uncomfortable procedure and might perhaps disturb the needle tip by so doing.