Commentary on “Postoperative complications and mechanical outcomes following endoscopic versus open lumbar fusion: a propensity-matched analysis”

Endoscopic lumbar fusion is moving rapidly from technical possibility to clinical practice. Wang et al. [1] provide a timely large-scale comparison of endoscopic and open lumbar fusion using propensity-matched national data [1]. Compared with open fusion, endoscopic fusion was associated with higher rates of several early complications and approximately twice the rate of pseudoarthrosis at both 1 and 2 years, yet with lower instrumentation failure and essentially identical revision rates. This discordant pattern challenges a basic question in fusion surgery: what constitutes success?

The early perioperative findings and the longer-term fusion-related outcomes warrant separate consideration. Regarding the former, the higher rates of urinary tract infection, pneumonia, hematoma, and surgical site infection observed after endoscopic fusion should be interpreted cautiously. Endoscopic fusion remains technically demanding, and at its current stage of adoption, differences in surgeon proficiency, operative efficiency, irrigation-dependent hemostasis, and constrained instrumentation may offset some of the advantages expected from reduced soft-tissue injury [2]. These limitations, however, are not necessarily intrinsic to the procedure, and further maturation of surgical technique, instrumentation, and workflow may allow the potential benefits of endoscopic fusion to be more fully realized.

The longer-term findings raise a different and arguably more fundamental issue. Interbody fusion requires adequate endplate preparation, sufficient bone grafting, broad graft–endplate contact, and mechanical stability, collectively creating a biological and mechanical environment conducive to new bone formation [3]. In endoscopic fusion, the narrower working corridor may make some of these requirements more difficult to achieve consistently. Although enhanced endoscopic visualization is an important advantage, visualization alone does not necessarily ensure complete preparation of the intended fusion surface. Likewise, accurate implant placement, less blood loss, and early symptom relief do not establish biological fusion.

The paradox of more pseudoarthrosis but less instrumentation failure may therefore be the most instructive observation in this study. Technical success, mechanical stability, biological fusion, and long-term durability are related, but they are not interchangeable. Instrumented spinal segments may remain mechanically stable despite incomplete arthrodesis, particularly in short-segment degenerative surgery with preserved posterior structures and percutaneous pedicle screw fixation. Conversely, the absence of implant failure does not prove fusion, just as radiographic nonunion does not necessarily require revision.

The critical variable is time. Instrumentation can compensate for incomplete biological fusion, but only while it continues to tolerate repetitive loading. A stable nonunion at 2 years may remain clinically silent, may eventually consolidate, or may represent a mechanical complication that has simply not yet declared itself. Long-term observations after conventional lumbar interbody fusion have shown that the radiographic and clinical course of pseudarthrosis may continue to evolve beyond the usual 1- to 2-year follow-up window [4]. Whether the same natural history applies to endoscopic interbody fusion remains unknown, because comparable long-term evidence is currently lacking. Similar 2-year revision rates therefore should not be interpreted as evidence of equivalent long-term durability.

The database itself imposes an equally important limitation. Pseudoarthrosis was identified from administrative diagnostic coding, without information on the method used to assess fusion or the detailed status of bone healing. Thus, the data do not permit a granular evaluation of fusion status or a clear distinction between delayed or incomplete fusion, established radiographic nonunion, and clinically symptomatic fusion failure. Important technical information—including graft type and amount, cage type and position, and the number and extent of fused levels—is also unavailable. Moreover, this was not specifically an elderly population, with a mean age of approximately 59 years. The findings may therefore not fully represent outcomes in older patients, in whom compromised bone quality and reduced biological capacity for fusion may become increasingly important. The reported twofold difference in pseudoarthrosis should accordingly be viewed as an important signal requiring further investigation rather than a definitive measurement of biological fusion failure.

This study ultimately asks us to reconsider the endpoint of endoscopic fusion. The goal of minimally invasive surgery is not simply to reproduce an operation through a smaller corridor. For a fusion procedure, success must ultimately include durable arthrodesis together with clinical benefit. Future studies should combine standardized CT-based fusion assessment, patient-reported outcomes, granular operative variables, and longer follow-up, particularly in older and osteoporotic populations.

A smaller corridor should complement, not compromise, the goal of durable fusion.

Funding

No specific funding was received for this work.

Declaration of competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.