Scholarlink

Purpose <p> To compare implant–acromion collision risk between the transdeltoid and anterolateral surgical approaches for straight antegrade humeral nailing (SAHN) using a CT-based simulation, and to evaluate whether the Acromion Index (AI) predicts collision.</p> Methods <p> Sixty-eight anonymized shoulder CTs were segmented to generate scapular and humeral surface models. A straight 8-mm nail was virtually advanced along the medullary axis under two conditions: (I) transdeltoid with 0° glenohumeral extension and (II) anterolateral with 30° extension. Two trained investigators independently repeated all simulations and recorded nail-acromion collisions. Agreement was assessed with Cohen’s κ and intraclass correlation coefficients (ICC). Collision rates were compared with paired McNemar tests. The predictive value of AI was analyzed using logistic regression and receiver-operating-characteristic (ROC) analysis.&#xa0;</p> Results <p> Implant-acromion collision occurred in 32/68 shoulders (47.1%) with the transdeltoid approach, versus 2/68 (2.9%) with the anterolateral approach (absolute risk reduction 44%; number-needed-to-treat 2.3; <i>p</i>&lt; 10⁻⁶). AI independently predicted transdeltoid failure (adjusted odds ratio 2.81 per +0.10; <i>p </i>= 0.011); a cut-off ≥ 0.69 yielded an AUC of 0.72 (63% sensitivity, 72% specificity). Inter-observer reliability was substantial for collision (κ 0.86) and excellent for morphometrics (ICC ≥ 0.90).&#xa0;</p> Conclusion <p> CT-based simulation demonstrates that the anterolateral approach markedly reduces the risk of implant-acromion collision in SAHN and should therefore be considered the preferred option. However, if the transdeltoid approach is selected, preoperative AI screening with a threshold of ≥0.69 is essential, as larger acromia increase collision risk and must be factored into surgical planning.&#xa0;</p> Level of evidence <p> Level IV – paired in-silico simulation study.</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

Novel CT-based simulation reveals anterolateral surgical approach avoids acromial collision in antegrade humeral nailing

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月13日

查看全文
  • Richard Arnhold,
  • Maximilian Kern,
  • Felix Schatzl,
  • Christoph Epple,
  • Franz Kralinger

摘要

Purpose

To compare implant–acromion collision risk between the transdeltoid and anterolateral surgical approaches for straight antegrade humeral nailing (SAHN) using a CT-based simulation, and to evaluate whether the Acromion Index (AI) predicts collision.

Methods

Sixty-eight anonymized shoulder CTs were segmented to generate scapular and humeral surface models. A straight 8-mm nail was virtually advanced along the medullary axis under two conditions: (I) transdeltoid with 0° glenohumeral extension and (II) anterolateral with 30° extension. Two trained investigators independently repeated all simulations and recorded nail-acromion collisions. Agreement was assessed with Cohen’s κ and intraclass correlation coefficients (ICC). Collision rates were compared with paired McNemar tests. The predictive value of AI was analyzed using logistic regression and receiver-operating-characteristic (ROC) analysis. 

Results

Implant-acromion collision occurred in 32/68 shoulders (47.1%) with the transdeltoid approach, versus 2/68 (2.9%) with the anterolateral approach (absolute risk reduction 44%; number-needed-to-treat 2.3; p< 10⁻⁶). AI independently predicted transdeltoid failure (adjusted odds ratio 2.81 per +0.10; p = 0.011); a cut-off ≥ 0.69 yielded an AUC of 0.72 (63% sensitivity, 72% specificity). Inter-observer reliability was substantial for collision (κ 0.86) and excellent for morphometrics (ICC ≥ 0.90). 

Conclusion

CT-based simulation demonstrates that the anterolateral approach markedly reduces the risk of implant-acromion collision in SAHN and should therefore be considered the preferred option. However, if the transdeltoid approach is selected, preoperative AI screening with a threshold of ≥0.69 is essential, as larger acromia increase collision risk and must be factored into surgical planning. 

Level of evidence

Level IV – paired in-silico simulation study.