Direct endoscopic visualization improves endplate preparation in lumbar interbody fusion: a comparative study of biportal endoscopic and conventional techniques.

PURPOSE: To compare the quality of endplate preparation between biportal endoscopic interbody fusion (BEIF) and conventional transforaminal lumbar interbody fusion (TLIF) in clinical cases using intraoperative endoscopic assessment.

METHODS: This retrospective study compared endplate preparation quality between BEIF (n = 25) and conventional TLIF (n = 22) in single-level L4-L5 lumbar fusion procedures. Endplate preparation quality was evaluated using standardized 30-degree endoscopic visualization with a 4-point grading system, assessing both ipsilateral and contralateral regions of upper and lower endplates. Inter- and intra-observer reliability of the grading system were assessed. Clinical outcomes were analyzed using linear mixed models, and fusion rates and complications were assessed at one-year follow-up.

RESULTS: BEIF demonstrated significantly better endplate preparation quality across all evaluated regions (all p < 0.001), with total scores of 10.76 ± 0.97 vs. 7.59 ± 1.37. Endplate violation rates were comparable between groups (upper: 20% vs. 27%, p = 0.732; lower: 36% vs. 45%, p = 0.562). Linear mixed model analysis revealed significantly better visual analog scale (VAS) back pain outcomes in the BEIF group (p = 0.011), which remained significant after baseline adjustment (adjusted p = 0.034). Both groups achieved comparable fusion rates at one-year follow-up (88% vs. 95%, p = 0.611).

CONCLUSIONS: BEIF may enable superior endplate preparation quality compared to conventional TLIF across all evaluated regions, and demonstrated significantly better postoperative back pain outcomes after controlling for baseline differences. However, given that the proposed grading system has not yet been externally validated, these results should be regarded as preliminary and hypothesis-generating.