The Legend of the Buffalo Chest.

BACKGROUND: The "buffalo chest" is a condition in which a simultaneous bilateral pneumothorax occurs due to a communication of both pleural cavities caused by an iatrogenic or idiopathic fenestration of the mediastinum. This rare condition is known by many clinicians because of a particular anecdote which stated that Native Americans could kill a North American bison with a single arrow in the chest by creating a simultaneous bilateral pneumothorax, due to the animal's peculiar anatomy in which there is one contiguous pleural space due to an incomplete mediastinum. RESEARCH QUESTION: What evidence is there for the existence of buffalo chest? STUDY DESIGN AND METHODS: The term "buffalo chest" and its anecdote were first mentioned in a ''personal communication'' by a veterinarian in the Annals of Surgery in 1984. A mixed method research was performed on buffalo chest and its etiology. A total of 47 cases of buffalo chest were identified in humans. RESULTS: This study found that all authors were referring to the article from 1984 or to each other. Evidence was found for interpleural communications in other mammal species, but no literature on the anatomy of the mediastinum of the bison was found. The main reason for this research was fact-checking the origin of the anecdote and search for evidence for the existence of buffalo chest. Autopsies were performed on eight bison, and four indeed were found to have had interpleural communications. INTERPRETATION: We hypothesize that humans can also have interpleural fenestrations, which can be diagnosed when a pneumothorax occurs.

Anatomical analysis of the left upper lobe of lung on three-dimensional images with focusing the branching pattern of the subsegmental veins.

BACKGROUND: A clear understanding of the anatomical characteristics of the pulmonary veins (PVs) is essential for the successful performance of segmentectomy and important to avoid intraoperative pulmonary vessels injury. However, there is no report showing the relations between the branching patterns of PVs and pulmonary arteries (PAs). Moreover, internationally accepted symbols for describing PVs remain unavailable. For anatomically assessing the branches and courses of the subsegmental veins in the left upper lobe (LUL), the diverse branching patterns of blood vessels and bronchi should be investigated. METHODS: The branching patterns and intersegmental courses of PVs were assessed by performing three-dimensional image analysis of the bronchi, and PAs and PVs in the LUL in 103 patients who were scheduled to receive segmentectomy in LUL from January 2008 through August 2012. RESULTS: Branching types of the bronchi and pulmonary vessels failed to be independent each other. Although the combinations of anterior extension type of bronchus with the inter-lobar type (IL-type) of arterial branching pattern were often observed, but those with the mediastinal type (M-type) were rarely observed. The combinations of apical vein dominant type with the IL-type of arteries, and intermediate and central vein types with the M-type were often observed. Since LUL was adjoined by various subsegments, and the intersegmental pulmonary veins showed diverse patterns. CONCLUSIONS: This study found the relationship among PA, PV, and bronchus patterns, in the subsegment where the branching patterns were fixed in 103 cases. This study discovered PVs that was difficult to be named by the conventional naming systems because of the diversity of the locations in the subsegment.

Predicting Successful Laparoendoscopic Transhiatal Esophagectomy (THE) by Mediastinal Height Measurement.

BACKGROUND: Laparoendoscopic transhiatal esophagectomy (THE) provides advantages over traditional THE by not only avoiding laparotomy but by also allowing more precise esophageal mobilization. Occasionally, the length of the gastric conduit is insufficient to allow delivery into the neck after laparoscopic mobilization and requires laparotomy to complete the procedure. We hypothesize that the need for laparotomy will correlate with the measurement of mediastinal height (distance from thoracic vertebrae T1-T12) on chest CT. METHODS: Medical records of all patients who underwent attempted laparoendoscopic-assisted THE at a tertiary referral center between March 1, 2003 and January 31, 2019 were reviewed. Patients' mediastinal height was measured using computed tomography (CT) imaging of the chest by investigators and analyzed for correlation between mediastinal height and successful completion of a totally laparoendoscopic procedure. RESULTS: A total of 21 cases met inclusion criteria: 9 successful laparoendoscopic THE procedures and 12 failed laparoendoscopic THE procedures (those requiring addition of a mini-laparotomy or thoracotomy). The mean mediastinal length for successful laparoendoscopic surgery was 23.5 cm, whereas the mean mediastinal length for failed laparoscopic surgeries was 24.8 cm (P = .03). Patient's overall height was not found to correlate with the need for conversion. CONCLUSIONS: Shorter mediastinal length is associated with successful laparoendoscopic or laparoscopic THE. This information is readily available to clinicians from routine preoperative staging studies (chest CT) and may be used to potentially predict the success rate of a totally laparoendoscopic approach and aid in patient selection. Further prospective evaluation of these findings is warranted.

Anomalies of the right vertebral vein increasing the difficulty of lymph-node dissection along the right recurrent laryngeal nerve: a single-institution, retrospective study.

BACKGROUND: Radical lymph-node dissection along the recurrent laryngeal nerves (RLN) improves the prognosis of patients with esophageal cancer. The RLN is a landmark for achieving adequate lymph-node dissection. However, the right RLN is sometimes covered by the right vertebral veins (VVs), making it undetectable. We investigated the relationship between this anomaly of the right VVs and the challenges of performing lymphadenectomy along the right RLN. METHODS: Patients with esophageal cancer, who underwent thoracoscopic esophagectomy with radical lymph-node dissection, were registered. The patterns of the right VVs were evaluated by preoperative computed tomography. The time required for identifying the right RLN or completing the lymphadenectomy was determined by reviewing surgical videos. RESULTS: In total, 178 patients were enrolled. Eighty patients (45%) had right VVs passing dorsal to the right subclavian artery (Dorsal group). More time was required to detect the right RLN in these cases (11 vs 9.5 min for the other cases, p = 0.034). In the Dorsal group, there were 15 patients who had specific VV patterns: The right VV converged on the lower portion of the right brachiocephalic vein (BCV), or passed through to the more medial side of the mediastinum. These patients required more time for detecting the right RLN (25 vs 9 min, p < 0.0001) and for completing the lymphadenectomy (41 vs 32 min, p = 0.048) than the other cases. CONCLUSION: The right VVs behind the subclavian artery, joining the lower part of the BCV or passing through the medial side, made it difficult to identify the right RLN and complete the lymphadenectomy.

Aortic inflation with agar injection is a useful method of cadaveric preparation which creates a mediastinal anatomy that better mimics the living body for surgical training.

In cadavers, even Thiel-embalmed cadavers, the arteries (especially the thoracic aorta) are extremely collapsed. This is in marked contrast to the state of the arteries in a living body. Aortic inflation is necessary to improve this unfavorable situation for anatomical observation or dissection. To inflate the aorta, we injected 500 ml of hot liquid agar into the aorta using a 18-Fr catheter inserted into the common femoral artery and subclavian artery. The injected agar then rapidly cools to room temperature and solidifies. As a result, the thoracic aorta remains sufficiently and constantly inflated in the mediastinum. This method is not only easy and inexpensive, but also useful and effective for achieving a life-like anatomy in cadavers used in surgical training for operations involving mediastinal organs, with the exception of the heart and great vessels.

[Thoracic lymphatic system-principles, characteristics and advice for the practice]. / Das thorakale Lymphsystem ­ Grundlagen, Besonderheiten und Hinweise für die Praxis.

The lymphatic system of the lungs is complex. To maintain an effective gas exchange there is a need for a dense lymphatic network. The alveolae have no lymphatic vessels. There is no segment-specific lymph drainage. For both lungs there are fixed bronchopulmonary lymph nodes but the number and size of the lymph nodes are variable. There are seven mediastinal lymph node chains that vary in extent, each of which acts as an independent functional unit. The accurate assessment of the nodal status needs a simple reproducible nodal map. The division into compartments or zones makes this easier. Mediastinal lymph node metastases without involvement of bronchopulmonary lymph nodes are possible. The development mechanism of this skip metastasizing is multifactorial.

CT-based mediastinal compartment classifications and differential diagnosis of mediastinal tumors.

Division of the mediastinum into compartments is used to help narrow down the differential diagnosis of mediastinal tumors, assess tumor growth, and plan biopsies and surgical procedures. There are several traditional mediastinal compartment classification systems based upon anatomical landmarks and lateral chest radiograph. Recently, the Japanese Association of Research of the Thymus (JART) and the International Thymic Malignancy Interest Group (ITMIG) proposed new mediastinal compartment classification systems based on transverse CT images. These CT-based classification systems are useful for more consistent and exact diagnosis of mediastinal tumors. In this article, we review these CT-based mediastinal compartment classifications in relation to the differential diagnosis of mediastinal tumors.

Comprehensive study of pulmonary hilam with its clinical correlation.

INTRODUCTION: Awareness of pulmonary hilar variations is essential for lobectomy of lung. MATERIALS AND METHODS: We studied 54 left and 49 right hilum of formalin fixed adult cadaveric lungs. Morphologic and mophometric details were recorded and variations were noted. RESULTS: Classical picture of hilum was found in 35.19% left lung and 40.82% right lung. Morphological variations were more on left side (64.81%) than right side (59.18%) in terms of numbers of structures. On the left side, highest percentage of variable structure was bronchus (46.3%) followed by pulmonary artery (37.31%) and lowest by pulmonary vein (31.48%) whereas on right side, percentage for variable pulmonary artery and vein were same (36.73%) followed by bronchi (20.41%). Maximum number of pulmonary veins was five, pulmonary artery was three and accessory bronchus was two on both side hila. In morphometric measurement, mean vertical length of hilum was more on right side whereas anteroposterior length was more on left side. Right hilum is slightly lower and anteriorly placed than left hilum in the mediastinal surface of lung. Significant correlations between vertical length of lung and hilum and antero-postero length of lung and hilum of left and right sides were found. CONCLUSIONS: By analysis and comparison with previous studies, present study concludes that morphology of pulmonary hila is extremely variable which contributes significant consequences in the field of pulmonary resection.

Fetal Congenital Pulmonary Airway Malformation: The Role of an Objective Measurement of Cardiomediastinal Shift.

OBJECTIVE: To examine the relationship between cardiomediastinal shift angle (CMSA) and adverse perinatal outcomes and hydrops in cases of congenital pulmonary airway malformation (CPAM). STUDY DESIGN: This retrospective study evaluated CPAM cases referred to our institution from 2008 to 2015. The primary outcome was a composite score for adverse perinatal outcome. CMSA was measured for each case and evaluated for its association with the primary outcome. The prediction accuracy of CMSA for adverse perinatal outcome was assessed using receiver operator characteristic (ROC) curves. RESULTS: Eighteen (21.2%) of the 85 cases experienced an adverse perinatal outcome. Increases in CMSA were associated with adverse perinatal outcomes and hydrops in bivariate analyses. Adjusted analyses found each 10-degree increase in CMSA to be associated with increased odds of an adverse perinatal outcome (adjusted odds ratio [aOR] 2.2, 95% confidence interval [CI]: 1.4-3.3) and hydrops (aOR 3.0, 95% CI: 1.5-6.1). CMSA performed well and was comparable to CPAM volume ratio in predicting adverse perinatal outcomes (area under the curve 0.81 and 0.84, respectively). CONCLUSION: We describe a novel measurement of mediastinal shift in cases of CPAM and its relationship with adverse perinatal outcomes and hydrops. These findings may shape the evaluation and management of CPAMs, improve our understanding of their prognosis, and influence patient counseling.

An anatomical hypothesis: a "concentric-structured model" for the theoretical understanding of the surgical anatomy in the upper mediastinum required for esophagectomy with radical mediastinal lymph node dissection.

Understanding the surgical anatomy is the key to reducing surgical invasiveness especially in the upper mediastinal dissection for esophageal cancer, which is supposed to have a significant impact on curability and morbidity. However, there is no theoretical recognition regarding the surgical anatomy required for esophagectomy, although the surgical anatomy in abdominal digestive surgery has been developed on the basis of embryological findings of intestinal rotation and fusion fascia. Therefore, we developed a hypothesis of a 'concentric-structured model' of the surgical anatomy in the upper mediastinum based on human embryonic development. This model was characterized by three factors: (1) a concentric and symmetric three-layer structure, (2) bilateral vascular distribution, and (3) an 'inter-layer potential space' composed of loose connective tissue. The concentric three-layer structure consists of the 'visceral layer', the 'vascular layer', and the 'parietal layer': the visceral layer containing the esophagus, trachea, and recurrent laryngeal nerves as the central core, the vascular layer of major blood vessels surrounding the visceral core to maintain the circulation, and the parietal layer as the outer frame of the body. The bilateral vascular distribution consists of the inferior thyroid arteries and bronchial arteries originating from the bilateral dorsal aortae in an embryo. This bilateral vascular distribution may be related to the formation of the proper mesentery of the esophagus and frequent lymph node metastasis observed in the visceral layer around recurrent laryngeal nerves. The three concentric layers are bordered by loose connective tissue called the 'inter-layer potential space'. This inter-layer potential space is the fundamental factor of our concentric-structured model as the appropriate surgical plane of dissection. The peripheral blood vessels, nerves, and lymphatics transition between each layer, thereby penetrating this loose connective tissue forming the inter-layer potential space. Recurrent laryngeal nerves also transition from the vascular layer after branching off from the vagal nerves and then ascend consistently in the visceral layer. We investigated the validity of this concentric-structured model, confirming the intraoperative images and the surgical outcomes of thoracoscopic esophagectomy in a prone position (TSEP) before and after the introduction of this hypothetical anatomy model. A total of 226 patients with esophageal cancer underwent TSEP from January 2015 to December 2016. After the introduction of this model, the surgical outcomes in 105 patients clearly improved for the operation time of the thoracoscopic procedure (160 min vs. 182 min, P = 0.01) and the incidence of recurrent laryngeal nerve palsy (19.0% vs. 36.4%, P = 0.004). Moreover, we were able to identify the concentric and symmetric layer structure through surgical dissection along the inter-layer potential space between the visceral and vascular layers ('viscero-vascular space') in all 105 cases after introduction of the hypothetical model. The concentric-structured model based on embryonic development is clinically beneficial for achieving less-invasive esophagectomy by ensuring a theoretical understanding of the surgical anatomy in the upper mediastinum.

Association of level of anastomosis and anastomotic leak after esophagectomy in anterior mediastinal reconstruction.

AIM: The purpose of this study was to investigate modifiable predisposing factors associated with anastomotic leak in the anterior mediastinal (AM) reconstruction route. METHODS: We reviewed the data on 154 patients who underwent esophagectomy and gastric tube reconstruction using the AM route between 2008 and 2016. The data included computed tomography (CT) scans with sagittal reconstruction of the thoracic section. The level of the esophagogastric anastomosis (LEA) and pretracheal distance (PTD) was measured from sagittal reconstructed CT images. Vascularization of the gastric tube was evaluated by postoperative endoscopy. Variables associated with anastomotic leak were determined using univariate and multivariate analyses. RESULTS: Anastomotic leak developed in 13 patients (8%). The cut-off level at which the anastomosis was less likely to develop a leak, as determined by Chi-square tests, was 1.5 cm for LEA and 1.3 cm for PTD. On univariate analysis, the factors that were significantly associated with the risk of anastomotic leak included diabetes, hand-sewn anastomosis, the LEA ≥ 1.5 cm, and severe mucosal degeneration. On multivariate analysis, diabetes (OR 4.7, 95% CI 1.29-17.2), LEA ≥ 1.5 cm (OR 20.1, 95% CI 3.15-128), and severe mucosal degeneration (OR 7.2, 95% CI 1.42-36.8) were found to be statistically significant independent risk factors. CONCLUSION: Use of the AM route to place the cervical anastomosis within 1.5 cm above the suprasternal notch might avoid excessive pressure on the gastric tube from the surrounding structures, resulting in a reduction in the risk of an anastomotic leak.

Histological study of the thin membranous structure made of dense connective tissue around the esophagus in the upper mediastinum.

BACKGROUND: The structure of the fascia in upper mediastinum has already been reported from gross anatomical viewpoints by Sarrazin. But it is necessary to understand meticulous anatomy for thoracoscopic or mediastinoscopic surgery. So herein, we investigate histologically the thin membranous structure made of dense connective tissues. METHODS: Semi-sequential transverse sections of the mediastinum were obtained from three cadavers. Hematoxylin and eosin staining, Elastica van Gieson staining, and Masson trichrome staining were performed to identify the presence and location of the thin membranous structure made of dense connective tissues. RESULTS: The "visceral sheath" and "vascular sheath," as previously described by Sarrazin, were observed histologically. These two thin membranous structures do not surround the esophagus and trachea cylindrically. In addition, the "visceral sheath" on the right side of the upper mediastinum was unclear in comparison to the left side. The "visceral sheath" (on the left side) gradually became unclear, and seemed to almost disappear; the esophagus was found to be very close to the thoracic duct on the caudal side of the bifurcation of the trachea. Although the left recurrent nerve was located inside the "visceral sheath" in all cadavers, the left recurrent nerve lymph nodes were located inside the "visceral sheath" in cadaver 1 and between the "visceral sheath" and "vascular sheath" in cadaver 3. CONCLUSION: The "visceral sheath" around the esophagus in the upper mediastinum was histologically demonstrated; however, the findings were not constant.

Plan optimization for mediastinal radiotherapy: Estimation of coronary arteries motion with ECG-gated cardiac imaging and creation of compensatory expansion margins.

BACKGROUND AND PURPOSE: Inadvertent heart and coronary arteries (CA) irradiation may increase the risk of coronary artery disease (CAD) in patients receiving thoracic irradiation. To date, the entity of cardiac-related CA displacement and the possible margins to be used for planning organs at risk volume (PRV) have been poorly described. Aim of this study was to quantify CA displacement and to estimate PRV through the use of ECG-gated computed tomography (CT) scans. MATERIAL AND METHODS: Eight patients received an ECG-gated intravenous contrast enhanced CT for non-cancer related reasons. Nine data sets were reconstructed over the entire R-R cycle with a dedicated retrospective algorithm and the following structures were delineated: Left main trunk (LM), left anterior descending (LAD), left circumflex (CX) and right coronary artery (RCA). CA displacements across the different cardiac phases were evaluated in left-right (X), cranio-caudal (Y) and anteroposterior (Z) directions using the McKenzie-van Herk formula (1.3 * Σ + 0.5 * σ). RESULTS: The following CA displacements were found in X, Y and Z coordinates: 3.6, 2.7 and 2.7 mm for LMT, respectively; 2.6, 5.0 and 6.8 mm for LAD, respectively; 3.5, 4.5 and 3.7 mm for CX, respectively; 3.6, 4.6 and 6.9 mm for RCA, respectively. Based on the mean displacements, we created a PRV of 3 mm for LM, 4 mm for CX and 5 mm for LAD and RCA. CONCLUSION: CA showed relevant displacements over the heart cycle, suggesting the need for a specific PRV margin to accurately estimate the dose received by these structures and optimize the planning process.

[Significance of the definition and detailing of the central neck compartment in the treatment of thyroid carcinoma].

The cervical lymph node metastasis is a risk factor for the high recurrence rate and low survival rate in patients with thyroid carcinoma (especially for age ≥45 years old), which is the most common head and neck carcinoma. The neck lymphatic metastasis, mostly occurring in the central neck compare, is diagnosed among 20%-90% of patients with differentiated thyroid carcinoma. To definite the anatomic boundaries of the central lymph nodes and their subgroups is significantly important for thyroid carcinoma, such as the determination of surgical extent, the evaluation of prognosis and the choice of different treatment strategies. This paper aims to describe the anatomic boundaries and the components of the central neck compartment and the role of individual subgroups in thyroid carcinoma. We think that the central neck compartment is bounded superiorly by the hyoid bone, inferiorly by the innominate artery on the right and the corresponding axial plane on the left, laterally by the carotid arteries. The superior mediastinal lymph nodes should be the important subgroups of the central neck compartment in thyroid carcinoma. When the prophylactic central lymph node dissection for involved lymph nodes is performed in thyroid carcinoma, we should pay more attention to the superior mediastinal lymph nodes.

Intrabronchial display of hilar-mediastinal lymph nodes by virtual bronchoscopic navigation system.

The description of precise intrabronchial positions for the sampling of mediastinal-hilar lymph nodes is critical to successfully perform conventional transbronchial needle aspiration. Previously published maps of mediastinal-hilar lymph nodes were primarily drawn based on experts' experience. We generated a virtual map of the most frequently sampled intrathoracic lymph nodes from an intrabronchial perspective using a virtual bronchoscopic navigation system, to assist with training in conventional transbronchial needle aspiration.

Clinical evaluation of atlas and deep learning based automatic contouring for lung cancer.

BACKGROUND AND PURPOSE: Contouring of organs at risk (OARs) is an important but time consuming part of radiotherapy treatment planning. The aim of this study was to investigate whether using institutional created software-generated contouring will save time if used as a starting point for manual OAR contouring for lung cancer patients. MATERIAL AND METHODS: Twenty CT scans of stage I-III NSCLC patients were used to compare user adjusted contours after an atlas-based and deep learning contour, against manual delineation. The lungs, esophagus, spinal cord, heart and mediastinum were contoured for this study. The time to perform the manual tasks was recorded. RESULTS: With a median time of 20 min for manual contouring, the total median time saved was 7.8 min when using atlas-based contouring and 10 min for deep learning contouring. Both atlas based and deep learning adjustment times were significantly lower than manual contouring time for all OARs except for the left lung and esophagus of the atlas based contouring. CONCLUSIONS: User adjustment of software generated contours is a viable strategy to reduce contouring time of OARs for lung radiotherapy while conforming to local clinical standards. In addition, deep learning contouring shows promising results compared to existing solutions.

Anatomical study of the left superior mediastinal lymphatics for tracheal branches of left recurrent laryngeal nerve-preserving mediastinoscope-assisted surgery in esophageal cancer.

PURPOSE: Curative treatment of esophageal cancer requires meticulous superior mediastinal lymphadenectomy, in addition to esophagectomy, because superior mediastinal lymph node metastases are common in esophageal cancer. When preserving the tracheal branches of the left recurrent laryngeal nerve (RLN), good anatomical understanding is required for confirmation of the positional relationships between the courses of lymphatic vessels, lymph node distribution, and the left RLN and its tracheal branches. We performed a detailed anatomical examination of these relationships. METHODS: Macroscopic anatomical observation and histological examination was performed on cadavers. In addition to hematoxylin and eosin staining, immunostaining using antipodoplanin antibody D2-40 (podoplanin) was performed to identify the lymphatic vessels. RESULTS: The tracheal branches of the left RLN were clearly observed, but no lymphatic vessels crossing the ventral or dorsal side of the branches were identified either macro-anatomically or histologically. CONCLUSION: No complex lymphatic network structure straddling the plane composed of tracheal branches of the left RLN was found in the left superior mediastinum. This suggests that dissection of the lymph nodes around the left RLN via the pneumomediastinum method using the left cervical approach may allow preservation of the tracheal branches of the left RLN by maintaining dissection accuracy.

Systematic intrafraction shifts of mediastinal lymph node targets between setup imaging and radiation treatment delivery in lung cancer patients.

BACKGROUND AND PURPOSE: Internal target motion results in geometrical uncertainties in lung cancer radiotherapy. In this study, we determined the intrafraction motion and baseline shifts of mediastinal lymph node (LN) targets between setup imaging and treatment delivery. MATERIAL AND METHODS: Ten lung cancer patients with 2-4 fiducial markers implanted in LN targets received intensity-modulated radiotherapy with a daily setup cone-beam CT (CBCT) scan used for online soft-tissue match on the primary tumor. At a total of 122 fractions, 5 Hz fluoroscopic kV images were acquired orthogonal to the MV treatment beam during treatment delivery. Offline, the 3D trajectory of the markers was determined from their projected trajectory in the CBCT projections and in the intra-treatment kV images. Baseline shifts and changes in the respiratory motion amplitude between CBCT and treatment delivery were determined from the 3D trajectories. RESULTS: Systematic mean LN baseline shifts of 2.2 mm in the cranial direction (standard deviation (SD): 1.8 mm) and 1.0 mm in the posterior direction (SD: 1.2 mm) occurred between CBCT imaging and treatment delivery. The mean motion amplitudes during CBCT and treatment delivery agreed within 0.2 mm in all directions. CONCLUSIONS: Systematic cranial and posterior intrafraction baseline shifts between CBCT and treatment delivery were observed for mediastinal LN targets. Intrafraction motion amplitudes were stable.

Fascial reinforcement fixing the bronchi to the heart: its anatomy and clinical significance.

PURPOSE: The details of the mediastinal fascia have been scarcely described and the bronchopericardial membrane is the only known structure that is present between the bronchi and the pericardium. However, the anatomical description of this structure is unclear. This study aimed to investigate the fascial structures between the bronchi and the pericardium based on surgical findings. METHODS: The connective tissues in the mid-mediastinum were observed surgically when lung lobectomy, including mediastinal lymph node dissection for lung cancer, was performed at our institute from April 2011 to March 2016. RESULTS: In total, 96 lobectomies were performed in 94 patients. A firm fibrous structure connecting the tracheobronchus and the fibrous pericardium was observed. It fixes the central bronchi to the pericardium and is composed of three parts. The largest part exists in front of the carina, its appearance is membranous, and runs behind the pulmonary artery. The other parts run over the right pulmonary artery and diverge at its superior trunk. The location at which all these structures fuse to the pericardium is the venous part of the hilum cordis (VHC). CONCLUSIONS: The results showed that connections of the dense fibrous tissues existed between the tracheobronchus and VHC. The structure not only works as a ligament that fixes the bronchi to the mid-mediastinum, but also divides the mid-mediastinum into two compartments: the Baréty and subcarinal spaces. The anatomy of the structure observed in this study differs from the previous description of the bronchopericardial membrane.