Letter to Editor Regarding “Comparative Efficacy and Safety of Minimally Invasive,Full-Endoscopic,and Biportal Transforaminal Lumbar Interbody Fusion for Lumbar Degenerative Diseases: A Network Meta-Analysis” by Cheng et al.

Dear Editor, we commend the authors for synthesizing the literature on endoscopic fusion, but we must highlight methodological concerns regarding the heterogeneity of studies and the premature nature of the comparison. Our recent review, ¹ focused on the differences between the Trans-Kambin and posterolateral endoscopic approaches, suggests that the conclusions may obscure important technical and biomechanical distinctions.

1 - Heterogeneity and different indications of “Full-Endoscopic” TLIF. The primary limitation of this network meta-analysis (NMA) is the aggregation of disparate surgical techniques under the term full-endoscopic-TLIF “FE-TLIF.” Uniportal endoscopic fusion is not a singular entity.^1,2 It involves the Trans-Kambin (KLIF) and the posterolateral (PE-TLIF) approaches. The KLIF uses a smaller working corridor, the Kambin triangle, and relies on indirect decompression. Historically, this approach has been limited by cage size, 7 to 10 mm wide, due to the proximity of the exiting nerve root. ³ Contrarily, the PE-TLIF (often used in FE or biportal endoscopy (BE)) involves facetectomy for direct decompression and insertion of larger cages. ⁴ The NMA combines techniques with different indications and risk profiles, potentially resulting in significant variations in subsidence and sagittal alignment correction rates between a small KLIF and a large PE-TLIF cage. ⁵

Also, the transitivity assumption is crucial for a valid NMA. It requires interventions to be jointly randomizable, which aren’t always held in clinical practice. KLIF is usually applied to “soft” issues like discogenic pain and mild spondylolisthesis, and contraindicated for severe central canal stenosis or calcified disc herniations. ⁶ Conversely, BE- and MIS-TLIF are commonly used in severe stenosis requiring extensive bony resection. ¹ Also, comparing “blood loss” or “hospital stay” between FE-TLIF and MIS-TLIF creates a selection bias favoring FE due to the lower complexity of cases in retrospective cohorts. ⁷ KLIF has limitations in cage size and sagittal alignment. The NMA suggests similar efficacy but overlooks these “Trans-Kambin” issues. ⁸ The constraints of Kambin’s triangle often require smaller cages, resulting in higher subsidence rates and undercorrection of PI-LL mismatch compared with larger cages used in MIS- or BE-TLIF. ⁶ As a result, short-term outcome comparisons overlook the potential long-term issues of undercorrected sagittal balance or pseudarthrosis.

2 - Insufficient High-Quality Comparisons and the Learning Curve. An NMA is premature given the scarcity of RCTs directly comparing FE-TLIF and BE-TLIF. This NMA is predominantly based on retrospective studies without control for the “learning curve” effect. ² FE-TLIF has a steep learning curve and higher operative times, which, in early-phase studies, bias the data against endoscopy. ² Complication rates may also be high due to proficiency issues rather than intrinsic technique flaws. The complication profile of KLIF (exiting nerve root dysesthesia) is distinct from and should be separated from the risks of dural tears associated with interlaminar approaches. ¹

In conclusion, while the NMA presents a rigorous summary of the data, its quality is insufficient to support broad claims of clinical equivalence. We advise caution when interpreting these findings, as “FE-TLIF” encompasses techniques with varying biomechanical effects. Future comparisons should differentiate between approaches and indications to provide meaningful guidance.