L3 translation predicts when L3 is not distal enough for an "ideal" result in Lenke 5 curves

Determining whether to fuse a Lenke 5 curve to L3 or to L4 is often a difficult decision. The purpose of this study was to determine preoperative variables predictive of an "ideal" or "less than ideal" outcome for Lenke 5 curves instrumented to L3.

A multicentre registry of adolescent idiopathic scoliosis patients was queried for surgically treated Lenke 5 curves with a lowest instrumented vertebra (LIV) of L3 and minimum 2 years of follow-up. Five seasoned surgeons qualitatively rated the 2-year postoperative images as "ideal" or "less than ideal" with respect to correction and alignment. Preoperative and postoperative radiographic variables were compared between the two groups. Multivariate regression analysis was performed to determine variables most predictive of a "less than ideal" outcome.

One hundred and thirty-nine patients met criteria. Twenty-three were considered "less than ideal" by ≥ 3 surgeons; 81 were unanimously "ideal". Preoperatively, the "less than ideal" group had significantly stiffer curves, greater apical translation, and greater LIV angulation and translation. Multivariate regression found that preoperative L3 translation (p = 0.009) was the single most important predictor of a "less than ideal" outcome: < 3.5 cm consistently resulted in an "ideal" outcome, while > 3.5 cm risked a "less than ideal" result.

While multiple variables are important in achieving an "ideal" outcome in Lenke 5 curves, this study found preoperative L3 translation was the most important predictor of success with an L3 translation < 3.5 cm being a potential threshold for selecting L3 as the LIV. These slides can be retrieved under Electronic Supplementary Material.