Abstract

Background: Spinal fusion surgery is challenging for patients aged ≥ 65 years. Minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) is the mainstay of treatment for degenerative diseases. We conducted this retrospective study to explore the role of MIS-TLIF in patients aged ≥ 65 years.

Methods: We included patients who underwent MIS-TLIF between January 2014 and July 2021. Clinical outcomes were evaluated according to pain severity, medication use, activity, and work status. Comorbidities, MIS-TLIF levels, operation duration, intraoperative blood loss, length of hospital stay, and postoperative complications were analyzed and compared between patients aged ≥ 65 years and those aged < 65 years. A predictive model for short-term clinical outcomes was constructed using decision tree-based machine learning.

Results: A total of 138 patients were enrolled. More than 86% of the patients had excellent and good short-term clinical outcomes. The predictive model had excellent accuracy, with the area-under-thecurve values being > 0.9. Risk factors associated with favorable short-term clinical outcomes were length of hospital stay ≥ 7 days (odds ratio [OR] = 0.36, p = 0.042), operation duration ≥ 240 minutes (OR = 0.295, p = 0.018), and MIS-TLIF levels (OR = 0.522, p = 0.04). However, multivariate analysis results revealed that the effects of these factors were nonsignificant. The approach-related complication rate in the patients who underwent MIS-TLIF was approximately 5%.

Conclusion: According to our highly accurate predictive model, MIS-TLIFwas effective in patients aged ≥ 65 years, with the corresponding short-term clinical outcomes being comparable to those observed in patient aged < 65 years.