A New Treatment Algorithm That Incorporates Minimally Invasive Surgery for Pyogenic Spondylodiscitis in the Thoracic and Lumbar Spines: The Results of Its Clinical Application to a Series of 34 Patients

Abstract

Background and Objectives: Spinal minimally invasive surgery (MIS) experts at the university hospital worked as a team to develop a new treatment algorithm for pyogenic spondylodiscitis in lumbar and thoracic spines. They modified a flow chart introduced for this condition in a pre-MIS era to incorporate MIS techniques based on their extensive experiences accumulated over the years, both in MIS for degenerative lumbar diseases and in the treatment of spine infections. The MIS procedures incorporated in this algorithm consisted of percutaneous pedicle screw (PPS)–rod fixation and transpsoas lateral lumbar interbody fusion (LLIF). The current study analyzed a series of 34 patients treated with prospective selection of the methods according to this new algorithm. Materials and Methods: The algorithm first divided the patients into those who had escaped complicated disease conditions, such as neurologic impairment, extensive bone destruction, and the need to be mobilized without delay (Group 1) (19), and those with complicated pyogenic spondylodiscitis (Group 2) (15). Group 1 had image-guided needle biopsy followed by conservative treatment alone with antibiotics and a spinal brace (12) (Group 1-A) or a subsequent addition of non-fused PPS–rod fixation (7) (Group 1-B). Group 2 underwent an immediate single-stage MIS with non-fused PPS–rod fixation followed by posterior exposure for decompression and debridement through a small midline incision (12) (Group 2-A) or an additional LLIF procedure after an interval of 3 weeks (3) (Group 2-B). Results: All patients, except four, who either died from causes unrelated to the spondylodiscitis (2) or became lost to follow up (2), were cured of infection with normalized CRP at an average follow up of 606 days (105–1522 days). A solid interbody fusion occurred at the affected vertebrae in 15 patients (50%). Of the patients in Group 2, all but two regained a nearly normal function. Despite concerns about non-fused PPS–rod instrumentation, only seven patients (21%) required implant removal or replacement. Conclusions: Non-fused PPS–rod placements into infection-free vertebrae alone or in combination with posterior debridement through a small incision worked effectively in providing local stabilization without contamination of the metal implant from the infected tissue. MIS LLIF allowed for direct access to the infected focus for bone grafting in cases of extensive vertebral body destruction.