Effect of segmental lordosis on early-onset adjacent-segment disease after posterior lumbar interbody fusion
Shinya Okuda, Yukitaka Nagamoto, Shota Takenaka, Masato Ikuta, Tomiya Matsumoto, Yoshifumi Takahashi, Masayuki Furuya, Motoki Iwasaki
Abstract
OBJECTIVE: Although several reports have described adjacent-segment disease (ASD) after posterior lumbar interbody fusion (PLIF), there have been only a few reports focusing on early-onset ASD occurring within 3 years after primary PLIF. The purpose of this study was to investigate the prevalence and postoperative pathologies of early-onset ASD and its relation with radiological parameters such as segmental lordosis (SL). METHODS: The authors reviewed a total of 256 patients who underwent single-segment PLIF at L4-5 for degenerative lumbar spondylolisthesis (DLS) and were followed up for at least 5 years. The definition of ASD was a symptomatic condition requiring an additional operation at the adjacent fusion segment in patients who had undergone PLIF. ASD occurring within 3 years after primary PLIF was categorized as early-onset ASD. As a control group, 54 age- and sex-matched patients who had not suffered from ASD for more than 10 years were selected from this series. RESULTS: There were 42 patients with ASD at the final follow-up. ASD prevalence rates at 3, 5, and 10 years postoperatively and at the final follow-up were 5.0%, 8.2%, 14.1%, and 16.4%, respectively. With respect to ASD pathologies, lumbar disc herniation (LDH) was significantly more common in early-onset ASD, while lumbar spinal stenosis and DLS occurred more frequently in late-onset ASD. Significant differences were detected in the overall postoperative range of motion (ROM) and in the changes in ROM (ΔROM) at L3-4 (the cranial adjacent fusion segment) and changes in SL (ΔSL) at L4-5 (the fused segment), while there were no significant differences in other pre- and postoperative parameters. In stepwise logistic regression analysis, ΔSL was identified as an independent variable (p = 0.008) that demonstrated significant differences, especially in early-onset ASD (control 1.1° vs overall ASD -2.4°, p = 0.002; control 1.1° vs early-onset ASD -6.6°, p = 0.00004). CONCLUSIONS: The study results indicated that LDH was significantly more common as a pathology in early-onset ASD and that ΔSL was a major risk factor for ASD, especially early-onset ASD.