Discitis after discography.

Infection after intradiscal injections has been recognised as a distinct entity, but discitis after discography has often been attributed to an aseptic process or a chemical reaction to the contrast material. We examined the hypothesis that discitis after discography is always due to infection, and report a clinical review and an experimental study. Part I. We reviewed the case records and radiographs of 432 patients who had undergone lumbar discography. When an 18-gauge needle without a stilette had been used, discitis was diagnosed in 2.7% of 222 patients but stiletted needles and a two-needle technique at each level reduced the incidence to 0.7%. Seven patients with discitis after discography had undergone anterior discectomy and fusion; in them the histopathological findings were of a chronic inflammatory response. Bacteria were isolated from the discs of three of the four patients who had open biopsy less than six weeks from the time of discography. These findings suggest that bacteria were initiators rather than promoters of the response. Part II. Multiple level lumbar discography was carried out in mature sheep, injecting contrast material with or without various concentrations of bacteria. Radiographs were taken and the discs and end-plates were examined histologically and cultured for bacteria at intervals after injection. None of the controls showed any evidence of discitis but all sheep injected with bacteria had typical radiological and histopathological changes by six weeks, though cultures were almost all negative. However, at one and two weeks after injection, but usually not after three weeks, bacteria could be isolated. We suggest that all cases of discitis after discography are initiated by infection, and that a very strict aseptic technique should be used for all injections into intervertebral discs.