Clinical and radiographic outcomes of robotic-assisted direct pars repairs for spondylolysis.

BACKGROUND: Spondylolysis is one of the most common identifiable causes of low back pain in young athletes. Direct pars repair offers a motion-preserving surgical option, and robotic-assisted minimally invasive surgery may reduce invasiveness and improve precision compared to traditional open techniques. METHODS: We performed a single-center retrospective review of patients with symptomatic spondylolysis who underwent robotic-assisted direct pars repair from 2023 to 2025. A reference array was placed on the spinous process of the affected vertebrae through a < 2 cm midline incision, and screws were placed through a separate < 2 cm midline incision. Hydroxyapatite screws were planned using pre- or intraoperative CT or synthetic CT. Screw position was confirmed with fluoroscopy and EMG. RESULTS: The study included 32 consecutive patients (mean 16 years, range 10-23) with 63 pars fractures (92% bilateral). Four patients (12%) had Grade 1 spondylolisthesis. Mean spondylolysis gap size was 1.5 mm (range 0.0-7.7 mm). Mean bilateral repair operative time was 118 min, and 27 patients (84%) were discharged same-day. Fourteen fractures had bone grafting at index surgery, and three patients (STIR negative) underwent bone grafting later for non-union. Four cases (13%) had minor perioperative complications with full resolution (anterior thigh pain/tightness, bilateral lower extremity paresthesia and transient pain that did not limit ambulation, seroma that resolved after bedside aspiration, guidewire breakage). There were no major complications or neurological injuries. At first postoperative follow-up (mean 21 days), 28 (88%) patients had complete resolution of pre-operative pain and 4 (12%) had improved pain. First CT (mean 10 weeks, n = 59 fractures) demonstrated complete union in 25 (42%), partial union in 24 (41%), and no union in 10 (17%). Among patients with delayed fracture healing requiring additional CT scans for surveillance (n = 28 fractures), final CT at a mean postoperative time of 27 weeks demonstrated complete union in 20 (71%), partial union in 3 (11%), and no union in 0 (0%). At final CT, STIR-positive fractures (n = 8) achieved 100% complete union versus 45% in STIR-negative fractures (n = 20). Screw placement matched preoperative planning in 100% of cases. One patient had a subsequent lamina fracture around the screw and will likely need revision. Twenty-nine of 32 patients (90.6%) returned to sport/baseline activity. CONCLUSION: This is the largest case series of spondylolysis treated with robotic-assisted direct pars repair. 100% of patients with a pre-operative STIR-positive MRI were pain-free and returned to sports. LEVEL OF EVIDENCE: level III.