Scholarlink

Purpose <p>Postoperative shoulder imbalance (PSI) in adolescent idiopathic scoliosis (AIS) surgery is associated with postoperative T1 tilt. Upper instrumented vertebra (UIV) tilt can be surgically adjusted; however, controlling T1 tilt remains challenging because the postoperative behavior of the unfused proximal thoracic curve (T1–UIV angle) is unpredictable. We aimed to develop and validate a regression-based equation for predicting the final follow-up (FFU) T1–UIV angle in AIS.</p> Methods <p>In this retrospective cohort study, we included 97 consecutive patients with AIS (Lenke Types 1 and 2) who underwent surgery. The T1–UIV angle was measured preoperatively, intraoperatively (post-correction), and at FFU. Factors associated with the FFU T1–UIV angle were evaluated using multiple linear regression based on predefined preoperative and surgically adjustable variables, and a regression-based equation was derived. Predictive performance was assessed using a separate internal validation cohort. Predicted T1 tilt, calculated as intraoperative UIV tilt + predicted FFU T1–UIV angle, was evaluated for its association with FFU radiographic shoulder height (RSH).</p> Results <p>Multivariate analysis identified the preoperative T1–UIV angle (B = 0.61, <i>p</i> &lt; 0.001) and UIV tilt (B = 0.08, <i>p</i> = 0.030) as independent factors associated with the FFU T1–UIV angle. The regression-based equation was defined as follows: Model 1 = 1.2 + 0.61 × (preoperative T1–UIV angle) + (preoperative UIV tilt / 12). To enhance clinical ease of use, a simplified single-parameter model was also developed: Model 2 = 0.6 × (preoperative T1–UIV angle). Bothmodels demonstrated good internal validity (Model 1: mean absolute error [MAE] = 2.07°, adjusted R² = 0.694; Model 2: MAE = 1.75°, adjustedR² = 0.757), with Model 2 showing comparable predictive performance despite its simplicity. Predicted T1 tilt was significantly associated with FFU RSH (<i>r</i> = 0.517, <i>p</i> &lt; 0.001).</p> Conclusion <p>These regression-based equations can provide a practical intraoperative reference for UIV tilt adjustment and may help reduce PSI. The simplified Model 2 offers a particularly practical bedside tool.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A regression-based equation for postoperative spontaneous correction of the unfused proximal thoracic curve in lenke types 1 and 2 adolescent idiopathic scoliosis

  • 期刊论文

  • 发表日期：

    2026年5月23日

查看全文
  • Masayoshi Iwamae,
  • Shinji Takahashi,
  • Yuki Kinoshita,
  • Minori Kato,
  • Hiromitsu Toyoda,
  • Akinobu Suzuki,
  • Koji Tamai,
  • Yuta Sawada,
  • Yuki Okamura,
  • Yuto Kobayashi,
  • Masato Uematsu,
  • Hiroshi Taniwaki,
  • Masashi Tsujino,
  • Hiroaki Nakamura,
  • Hidetomi Terai

摘要

Purpose

Postoperative shoulder imbalance (PSI) in adolescent idiopathic scoliosis (AIS) surgery is associated with postoperative T1 tilt. Upper instrumented vertebra (UIV) tilt can be surgically adjusted; however, controlling T1 tilt remains challenging because the postoperative behavior of the unfused proximal thoracic curve (T1–UIV angle) is unpredictable. We aimed to develop and validate a regression-based equation for predicting the final follow-up (FFU) T1–UIV angle in AIS.

Methods

In this retrospective cohort study, we included 97 consecutive patients with AIS (Lenke Types 1 and 2) who underwent surgery. The T1–UIV angle was measured preoperatively, intraoperatively (post-correction), and at FFU. Factors associated with the FFU T1–UIV angle were evaluated using multiple linear regression based on predefined preoperative and surgically adjustable variables, and a regression-based equation was derived. Predictive performance was assessed using a separate internal validation cohort. Predicted T1 tilt, calculated as intraoperative UIV tilt + predicted FFU T1–UIV angle, was evaluated for its association with FFU radiographic shoulder height (RSH).

Results

Multivariate analysis identified the preoperative T1–UIV angle (B = 0.61, p < 0.001) and UIV tilt (B = 0.08, p = 0.030) as independent factors associated with the FFU T1–UIV angle. The regression-based equation was defined as follows: Model 1 = 1.2 + 0.61 × (preoperative T1–UIV angle) + (preoperative UIV tilt / 12). To enhance clinical ease of use, a simplified single-parameter model was also developed: Model 2 = 0.6 × (preoperative T1–UIV angle). Bothmodels demonstrated good internal validity (Model 1: mean absolute error [MAE] = 2.07°, adjusted R² = 0.694; Model 2: MAE = 1.75°, adjustedR² = 0.757), with Model 2 showing comparable predictive performance despite its simplicity. Predicted T1 tilt was significantly associated with FFU RSH (r = 0.517, p < 0.001).

Conclusion

These regression-based equations can provide a practical intraoperative reference for UIV tilt adjustment and may help reduce PSI. The simplified Model 2 offers a particularly practical bedside tool.