Photo by rossfindon from unsplash
W e carefully read the article published in Spine by Kato et al. The work was with great effort. However, in the study, the surgical criteria for the selection of… Click to show full abstract
W e carefully read the article published in Spine by Kato et al. The work was with great effort. However, in the study, the surgical criteria for the selection of lowest instrumented vertebra (LIV) were not discussed, which we believed was crucial to the L3-4 disc wedging angle postoperatively and at final follow-up time. As 53 patients (68%) had structural thoracolumbar curvature, the change of L3-4 disc wedging angle was totally different between the patients with lower end vertebra (LEV) at L3 and L4. Therefore, we have some concerns about this article. Lee et al indicated stopping fusion at L3 may have different adjacent disc wedge angle, correction rate and change of apical vertebral translation in LEV 4 and LEV 3 patients. We think the authors could report the clear surgical criteria for the patients fusion to L3 and show the outcome of postoperative distal coronal decompensation in patients with different LEV. In this study, the authors found interesting association between increased pelvic tilt and increased L3-4 disc wedging. However, the case example in this study had a severe proximal junctional kyphosis and the Cobb angle of lumbar curve was much >438. The relationship between the change of pelvic tilt (PT), proximal junctional kyphosis, and the selection of LIV may influence the final result. Thus, we thought the change of proximal junctional kyphosis after surgery should be taken into calculated in this study instead of only discussing the lumbar-pelvic parameters At last, LIV translation, Cobb angle, pelvic incidence, and PT were used in the multivariate logistic regression analysis, which revealed that larger PT was a significant risk factor for decompensation. However, Shu et al’s study also showed the immediate postoperative LIV tilt and LIVDA may also influence the final adding-on. Why not take the LIV tilt and the immediate postoperative parameters in the multivariate logistic regression analysis and get a more predictable result? We wish the authors to show more data about these questions.