LAUSR

Abstract Background: The overall benefit of surgical management in the acute phase of hematogenous pyogenic vertebral osteomyelitis remains difficult to evaluate because of the balance between potential functional benefit versus complications of surgery. Methods: Between 2000 and 2013, in a tertiary care hospital, we analyzed a cohort of patients with hematogenous pyogenic vertebral osteomyelitis treated surgically and compared them to those treated medically. Neurologic deficit (using the ASIA impairment scale) and pain (using the analgesic level required) 4 months later, recurrences and infection-related deaths 12 months later were evaluated. A propensity score was developed to adjust for nonrandomized allocation to surgery. Results: Ninety patients were included (mean age 64 years, 63% male); 28 (31%) were treated surgically. After adjustment for the propensity score, the improvement in neurological deficit at 4 months did not differ between surgical and medical treatment (p = .82), but the reduction of pain tended to be greater in surgical versus medical treatment (p = .051). Recurrences of infection (5%) and infection-related deaths (12%) occurred at similar rates in both groups at 12 months (p = 1.00 for both). Conclusions: Patients with hematogenous pyogenic vertebral osteomyelitis requiring surgery improved equally as non-surgical patients in terms of neurological deficit and residual pain. This result was not hampered by increased complications related to surgery. When indicated, surgery is safe and effective in patients suffering from hematogenous pyogenic vertebral osteomyelitis.