Abstract
Diagnostic ultrasound has mainly been utilised in the field of back pain for the detection of lumbar canal stenosis and the identification of visceral pathologies, with associated referred lumbar pain. Following the published accuracy of lumbar canal measurement using a compound scanner, it was postulated that the instability of the neural arch, in patients with spondylolysis, could be objectively demonstrated. A technique was designed for the detection of this instability using a compound ultrasound scanner. A pilot study indicated that the method was able to distinguish between affected and unaffected patients. The procedure was, however, time consuming, the equipment expensive and being non-portable made preventative screening in school children impractical. The suitability of different scanners for lumbar canal measurement was thus assessed. A method for measuring the lumbar canal, using a portable realtime scanner was developed. Subsequent investigations to detect spondylolysis, however, gave errors greater than the variations expected in an affected subject. A number of in vitro experiments were therefore designed to determine the geometrical parameters, sources of error and origins of echoes during lumbar sonography. Echoes were found to come from both the bony and soft tissue elements of the spine. Pulsations were visualised from the soft tissues within the spinal canal though these were dependent on scanning plane. The precision of the ultrasound system was investigated and considered not to be a limiting factor in the success of this technique providing suitable equipment was used. The major sources of error were considered to be biological or due to operator error. The literature relative to spondylolysis and lumbar sonography was reviewed. The physical principles of pulse-echo ultrasound were discussed especially in relation to lumbar sonography. Other possible uses of diagnostic ultrasound in the field of back pain were briefly investigated.