Background and study aims:

Spinal instrumentation for spondylodiskitis (SD) remains highly controversial. To date, surgical data are limited to relatively small case series with short-term follow-up data. In this study, we wanted to elucidate the biomechanical, surgical, and neurologic long-term outcomes in these patients.

Material and methods:

A retrospective analysis from two German primary care hospitals over a 9-year period (2005-2014) was performed. The inclusion criteria were (1) pyogenic lumbar SD, (2) minimum follow-up of 1 year, and (3) surgical instrumentation. The clinical and radiologic outcome was assessed before surgery, at discharge, and at a minimum of 12 months of follow-up. Follow-up included physical examination, laboratory results, CT and MRI scans, as well as assessment of quality of life (QoL) using short-form health survey (SF-36) inventory, Oswestry Disability Questionnaire, and visual analog scale (VAS) spine score.

Results:

Complete data were available in 70 patients (49 males and 21 females, with an age range of 67±12.3 years) with a median follow-up of 6.6 ± 4.2 years. Follow-up data were available in 70 patients after 1 year, in 58 patients after 2 years, and in 44 patients after 6 years. Thirty-five patients underwent posterior stabilization and decompression alone and 35 patients were operated on in a two-stage 360-degree interbody fusion with decompression. Pre- and postoperative angles of the affected motion segment were 17.6 ± 10.2 and 16.1 ± 10.7 degrees in patients with posterior instrumentation only and 21.0 ± 10.2 and 18.3 ± 10.5 degrees in patients with combined anterior/posterior fusion. Vertebral body subsidence was seen in 12 and 6 cases following posterior instrumentation and 360-degree instrumentation, respectively. Nonfusion was encountered in 22 and 11 cases following posterior instrumentation and 360-degree instrumentation, respectively. The length of hospital stay was 35.0 ± 24.5 days. Surgery-associated complication rate was 18% (12/70). New neurologic symptoms occurred in 7% (5/70). Revision surgery was performed in 3% (2/70) due to screw misplacement/hardware failure and in 3% (2/70) due to intraspinal hematoma. Although patients reported a highly impaired pain deception and vitality, physical mobility was unaffected and pain disability during daily activities was moderate.

Conclusion:

Surgical treatment of SD with a staged surgical approach (if needed) is safe and provides very good long-term clinical and radiologic outcome.