|Home | About | Journals | Submit | Contact Us | Français|
Abscess formation, from deep fissuring of the bowel wall or suppuration within lymph nodes and local fistulation, develops in as many as 20% of patients with Crohn’s disease.1 Several cases of vertebral and paravertebral infection have been described in association with enteric fistulas. Isolated discitis, however, has been little reported.
A man of 58 consulted his general practitioner because of abdominal pain, diarrhoea and weight loss. An abdominal X-ray had shown partial obstruction of the large bowel and the possibility of air within the bowel wall. After admission to hospital he developed peritonitis from perforation of the descending colon and underwent left hemicolectomy, loop ileostomy and splenectomy (the inflamed colon being adherent to the spleen). Postoperatively he did well until the onset of high fever and leucocytosis due to an intra-abdominal collection. Percutanous drainage of this collection resulted in a chronic colocutanous fistula, specimens from which grew pseudomonas species. The fistula improved and he was discharged from hospital. A few weeks later, however, he had to be readmitted with recurrence of his intra-abdominal collection and the fistula. He also had severe back pain. Again the fistula grew pseudomonas species. A sinogram showed a possible connection to bowel at the colonic anastomosis, and an incidental finding was spondylolisthesis of the fifth lumbar and first sacral vertebrae (L5/S1).
After several weeks of severe back pain MRI of his lower back showed inflammation in the L5/S1 disc but no connection between the disc and the fistula (Figure 1 and Figure 2). Aspiration of the disc yielded pus from which pseudomonas species were grown. The patient was treated with long-term antibiotics and became free from back pain. The fistula also healed. Subsequent colonoscopy revealed inflammation of the transverse and ascending colon, and biopsy appearances were compatible with Crohn’s disease. He was started on azathioprine and the ileostomy was closed without incident.
The striking feature of this case was the apparent lack of a fistula from the gut to the infected disc. Pande and co-workers2 described a patient not unlike ours with a presacral abscess and osteomyelitis of the L4 and L5 vertebrae, but the route of infection in that case was clearly an enteric fistula. We postulate that in our patient the route of infection was haematogenous, through the vertebral venous system and then retrogradely into the vertebral venous plexus.