Not all Parkinsons patients with thoracolumbar spinal fusion are created equal: Highlighting the difference between...

Parkinsons disease (PD) is a devastating neuromuscular disease that has several distinct spinal syndromes in addition to common degenerative spinal disorders. Multiple studies have concluded that patients with Parkinson's disease are at increased risk for perioperative medical complications and poor surgical outcomes such as failure of hardware and reoperation. PD patients exist in a spectrum of spinal disease with presentations, indications and surgical paradigms varying widely, yet the literature tends to treat them as a homogenous group. A more nuanced investigation and comparison of PD patients undergoing TL fusion will yield important differences in presentation, indication and outcomes. There is scant evidence on the most appropriate fusion paradigm for treating Parkinson's patients. We set out to compare patients who had short or long (>3 levels) thoracolumbar fusions to contrast the rate of surgical failure, and complications between each surgical approach and identify any patient, surgical, or disease risk factors for poor outcomes.

This retrospective cohort study of a single center experience with spinal fusion patients with concomitant Parkinson's disease at the time of spinal surgery. We included adult patients who underwent thoracolumbar spinal fusion and had a concomitant diagnosis of PD at an academic center between 2017 and 2022. Revision, oncologic and infection cases were excluded. Primary outcome measures were radiographic (stenosis, listhesis, hardware, fracture) evidence of failure, and morbidity. Secondarily, a multivariate regression was performed to identify patient risk factors for failure. We extracted and analyzed demographic information, Parkinson's specific metrics, in and out of hospital complications, surgical data, and radiologic outcomes and compared patients who underwent either short or long fusion constructs.

92 patients were identified after inclusion and exclusion. 63 (68%) underwent short fusion and 29 (32%) had long fusion constructs. Short fusion constructs were more strongly indicated for radiculopathy 51% vs. 21% p=.01) while extended fusions had a higher indication of spinal deformity (41%vs. 10% p=.001) or fracture (34%vs. 8% p=.004) and were more likely to have decreased mobility prior to surgery (79.7%vs. 53.9%; P<.001). Short constructs had a higher rate of listhesis (20%vs. 0% p=.01) or stenosis (34.5%vs. 4% p=.008) at the level above the upper instrumented level, proximal screw loosening (13%vs. 0% p=.009), proximal junctional complications (53%vs. 12% p=.001) and proximal junctional failures (37%vs. 8% p=.02). In hospital complications, 90 day and 1 year mortality, reoperation rate, infection rate, progressive neurologic decline, and fusion rates were statistically similar between groups.

How to approach surgical treatment of PD patients remains a clinical conundrum. This study compared PD patients who had short and long constructs and showed statistical equipoise in mortality, complications and re-operations between the 2 cohorts. While differing in presentation, preoperative debility and indications, both short and long fusions in PD patients have elevated but acceptable complication rates. When long fusion is properly indicated, it can be performed without major increases in complications. We found that short fusions have relatively high rates of screw loosening and de novo adjacent level complications.