Predicting life-threatening hemoptysis in traumatic pulmonary parenchymal injury using computed tomography semi-automated lung volume quantification

Wen-Ruei Tang; Chao-Chun Chang; Chen-Yu Wu; Chih-Jung Wang; Tsung-Han Yang; Kuo-Shu Hung; Yi-Sheng Liu; Chia-Ying Lin; Yi-Ting Yen

doi:10.1186/s13244-024-01849-8

Predicting life-threatening hemoptysis in traumatic pulmonary parenchymal injury using computed tomography semi-automated lung volume quantification

Insights Imaging. 2024 Nov 15;15(1):276. doi: 10.1186/s13244-024-01849-8.

Authors

Wen-Ruei Tang¹, Chao-Chun Chang^{1

2}, Chen-Yu Wu^{1

2}, Chih-Jung Wang², Tsung-Han Yang², Kuo-Shu Hung², Yi-Sheng Liu³, Chia-Ying Lin⁴, Yi-Ting Yen^{1

2}

Affiliations

¹ Division of Thoracic Surgery, Department of Surgery, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan.
² Division of Trauma and Acute Care Surgery, Department of Surgery, National Cheng Kung University Hospital, Tainan, Taiwan.
³ Department of Medical Imaging, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan.
⁴ Department of Medical Imaging, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan. gracelinchiaying@msn.com.

Abstract

Objectives: Chest computed tomography (CT) can diagnose and assess the severity of pulmonary contusions. However, in cases of severe lung contusion, the total lung volume ratio may not accurately predict severity. This study investigated the association between life-threatening hemoptysis and chest CT imaging data on arrival at the emergency department in patients with pulmonary contusions or lacerations due to blunt chest injury.

Methods: The records of 277 patients with lung contusions or lacerations treated at a trauma center between 2018 and 2022 were retrospectively reviewed. The ratio of the local lung contusion volume to lobe volume in each lobe was calculated from chest CT images. The maximal ratio in the Hounsfield unit (HU) range was defined as the highest ratio value within the HU range among five lobes.

Results: The median patient age was 41 years, and 68.6% were male. Life-threatening hemoptysis occurred in 39 patients. The area under the receiver operating characteristic curve for the maximal ratio at -500 HU to 100 HU was 96.52%. The cutoff value was 45.49%. Multivariate analysis showed a high maximal chest CT ratio ≥ 45.49% at -500 HU to 100 HU (adjusted odds ratio [aOR]: 104.66, 95% confidence interval [CI]: 21.81-502.16, p < 0.001), hemopneumothorax (aOR: 5.18, 95% CI: 1.25-21.47, p = 0.023), and chest abbreviated injury scale (AIS, aOR: 5.58, 95% CI: 1.68-18.57, p = 0.005) were associated with life-threatening hemoptysis.

Conclusions: Maximal chest CT ratios ≥ 45.49% at -500 HU to 100 HU, hemopneumothorax, and high chest AIS scores are associated with life-threatening hemoptysis in patients with blunt chest trauma.

Critical relevance statement: The present study provides an objective index derived from chest CT images to predict the occurrence of life-threatening hemoptysis. This information helps screen high-risk patients in need of more intensive monitoring for early intervention to improve outcomes.

Key points: Emergency department CT helps predict life-threatening hemoptysis in patients with lung contusions. Maximal CT ratios ≥ 45.49% (-500 HU to 100 HU, either lung lobe) are associated with life-threatening hemoptysis. High chest abbreviated injury scale scores and hemopneumothorax also predict life-threatening hemoptysis.

Keywords: Computed tomography; Hemoptysis; Pulmonary contusion; Quantitative CT analysis; Thoracic trauma.