Introduction <p>Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is vital for diagnosing mediastinal/hilar lymphadenopathy. This study investigated the diagnostic performance of Hounsfield unit (HU) values and PET-CT SUVmax in differentiating malignant from benign lymph nodes.</p> Methods <p>This retrospective analysis included 168 patients who underwent EBUS-TBNA between 2021 and 2024. Data on demographics, smoking history, HU, PET-CT SUVmax, and EBUS lymph node measurements were collected. Statistical methods included logistic regression and receiver operating characteristic (ROC) analysis to assess diagnostic accuracy.</p> Results <p>Malignant lymph nodes had significantly lower HU values and larger EBUS dimensions (<i>p</i> &lt; 0.05). The PET‒CT SUVmax did not significantly differ (<i>p</i> = 0.607). ROC analysis revealed that HU had the highest predictive accuracy (AUC = 0.678). Logistic regression identified HU as an independent predictor of malignancy.</p> Conclusion <p>CT-based HU measurements have greater diagnostic value than PET-CT SUVmax in predicting malignancy in EBUS-TBNA-biopsied lymph nodes. HU could aid in guiding biopsy decisions and improve diagnostic accuracy.</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

Underestimated value of Hounsfield unit in differentiating benign and malignant lymph nodes via EBUS-TBNA

  • Caner Çınar,
  • Zeynep Mercancı,
  • Şehnaz Olgun Yıldızeli,
  • Derya Kocakaya,
  • Emine Bozkurtlar,
  • Sait Karakurt

摘要

Introduction

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is vital for diagnosing mediastinal/hilar lymphadenopathy. This study investigated the diagnostic performance of Hounsfield unit (HU) values and PET-CT SUVmax in differentiating malignant from benign lymph nodes.

Methods

This retrospective analysis included 168 patients who underwent EBUS-TBNA between 2021 and 2024. Data on demographics, smoking history, HU, PET-CT SUVmax, and EBUS lymph node measurements were collected. Statistical methods included logistic regression and receiver operating characteristic (ROC) analysis to assess diagnostic accuracy.

Results

Malignant lymph nodes had significantly lower HU values and larger EBUS dimensions (p < 0.05). The PET‒CT SUVmax did not significantly differ (p = 0.607). ROC analysis revealed that HU had the highest predictive accuracy (AUC = 0.678). Logistic regression identified HU as an independent predictor of malignancy.

Conclusion

CT-based HU measurements have greater diagnostic value than PET-CT SUVmax in predicting malignancy in EBUS-TBNA-biopsied lymph nodes. HU could aid in guiding biopsy decisions and improve diagnostic accuracy.