Introduction
For thoracic surgeons, knowledge of pulmonary bronchovascular patterns, including rare anatomical variations, is extremely important to perform safe and accurate anatomical pulmonary resections. However, currently available anatomical data on pulmonary bronchovascular patterns are limited to a few cadaveric studies. These studies include the small number of cases reported by Boyden et al. [
1,
2] and Yamashita [
3], performed between the 1950s and 1970s. In a previous study, we focused on the usefulness of 3DCT imaging to understand individual variation in thoracic anatomy, and previously reported a systematic radiological analysis of the pulmonary structure of the right upper lobe (RUL) using the largest number of cases in the literature (
n = 263) [
4]. In this study, we analyzed variation in the pulmonary bronchovascular pattern using 3DCT angiography and bronchography (3DCTAB) of the right middle lobe (RML) and right lower lobe (RLL) of the lung.
Patients and methods
Reconstruction of 3DCTAB imaging
Bronchovascular patterns revealed by 3DCTAB imaging were analyzed by 64-channel multidetector row computed tomography (MDCT) (SOMATOM Definition Flash; Siemens Healthcare, Berlin, Germany). A total of 35 mL contrast agent was mechanically injected at 5 mL/s, immediately followed by injection of 20 mL of saline. A solid image was constructed from 1.0-mm data slices of contrast-enhanced CT images with the aid of 3D volume rendering. The volume data from both arterial and venous phases were transferred to a workstation running volume-rendering reconstruction software (Ziostation2; Ziosoft, Tokyo, Japan) that converted the data into the 3DCT angiographic format. The 3D reconstruction of the bronchial tree involved mathematical morphology-based 2D segmentation of axial images, followed by restoration via manual addition of segments from 2D axial images to form 3D images. Radiology technicians processed all 3D images, and thoracic surgeons confirmed the validity of the reconstructed images. The detection rate of pulmonary vessels by our 3DCTAB was 98.7% [
4], indicating the feasibility of comparing our 3DCTAB data to previously reported data based on actual anatomy.
Patient preparation and examination
Between November 2010 and November 2015, 302 patients with respiratory or mediastinal lesions underwent 3DCTAB prior to surgery. Thirty-two cases were excluded because some subsegmental branches of the pulmonary vessels and bronchi were not adequately represented on 3DCTAB due to technical issues, such as obstruction by lung tumors and/or lymph node metastasis. Therefore, we analyzed variation in the pulmonary vessel patterns of the most recent 270 consecutive cases (150 men, 120 women; median age 67 years) in whom all subsegmental branches were properly represented on 3DCTAB. Patient characteristics are summarized in Supplemental Table 1.
The frequencies of each bronchovascular pattern in our study and those of previous studies (Boyden et al. [
1,
2] and Yamashita [
3]) were compared using the
χ
2 test. However, some branching patterns were not described in these previous reports [
1‐
3], and thus our data could not compare for those patterns. All statistical analyses were performed using SPSS ver. 22 software (SPSS, Inc., Chicago, IL, USA). This study was approved by the Research Ethics Committee at Gunma University Hospital, Maebashi, Gunma, Japan.
Definition of pulmonary vessels and bronchi
We used the same nomenclature to describe segmental structures used by Boyden et al. [
1,
2] and Yamashita [
3]. Segmental and subsegmental vessels were named with reference to their relationships to the segmental bronchi.
We classified the branching patterns of vessels and bronchi in the RML and S
6 according to the number of stems. However, the bronchovascular pattern of the right basal segments (S7, S8, S9, and S10) was more complex than those of the RML and S
6. Therefore, we defined the pattern of the right basal segment according to Yamashita’s [
3] classification as follows:
Subsuperior segment (S*) The independent segment that is infrequently observed between S6 and S10 is called the subsuperior segment, S* (Suppl Fig. 1). S* has an independent segmental bronchus (B*) and artery (A*) that are distinct from those of S6 and S10. B* (A*) bifurcates from between the basal bronchus (artery) and B10 (A10), and B* (A*) points in a posterolateral direction directly toward the vertebral body.
Mediobasal segment (S
7
) A7 (B7) divides into an anterior ramus A7a (B7a) and a posterior ramus A7b (B7b). A7a (B7a) runs anterior to the inferior pulmonary vein (IPV), whereas A7b (B7b) runs posterior to the IPV. A7 (B7) is classified into four types according to the combination of A7a (B7a) and A7b (B7b). The A7a (B7a) type has only A7a (B7a), and in this type, S7 is located anterior to IPV. The A7b (B7b) type has only A7b (B7b), and in this type, S7 is located posterior to IPV. The A7ab (B7ab) type has both A7a (B7a) and A7b (B7b), and in this type, S7 is located both anterior and posterior to IPV. Lack of an original A7 (B7) was occasionally observed. In such instances, S7 was supplied by an accessory artery (bronchus) branching from A8 (B8), A9 (B9), or A10 (B10). This type was named the AX7 (BX7) type. B7 was classified in accordance with the artery.
Ventrobasal, laterobasal, and dorsobasal segments (S
8
, S
9
, S
10
) The branching patterns of the ventrobasal, laterobasal, and dorsobasal segmental arteries and veins were classified into two types: bifurcation and trifurcation types. The bifurcation type was further divided into two subtypes: simple and split. In the simple bifurcation type, each basal segment was supplied by a single segmental artery: A8 and A9 + A10 type and A8 + A9 and A10 type. In the split type, S8 or S9 was supplied by two segmental arteries: A8 and A8 + A9 + A10 type and A8 + A9 and A9 + A10 type. The ventrobasal, laterobasal, and dorsobasal segmental veins were also classified according to the artery. In the simple type, each basal segment was drained by a single segmental vein: V8 and V9 + V10 type and V8 + V9 and V10 type. In the split type, S9 or S10 was drained by two segmental veins: V8 + V9 and V9 + V10 type and V8 + V9 + V10 and V10 type. The ventrobasal, laterobasal, and dorsobasal segmental bronchi were also classified in accordance with vessels. However, there were no split bifurcation types.
Discussion
We performed systematic radiological analyses of the pulmonary structure of the RML and RLL, and examined the discrepancies between our results and those of previous reports [
1‐
3]. Only a few reports have focused on thoracic anatomical abnormalities detected with the aid of 3DCT [
5,
6]. However, there have been no previous systematic explorations of the bronchovascular pattern using 3DCT imagery. Therefore, this is the first systematic radiological analysis of the pulmonary structure of the RML and RLL. Moreover, our study included a larger number of cases (
n = 270) than all previous relevant reports.
The frequencies of some branching types differed from those described in previous reports [
1‐
3], and these discrepancies were particularly evident for the basal segment. Moreover, the bronchovascular pattern of S
7 was more complex than that of S
8−10 because the branching pattern of B
7 (and A
7) depends on its relationship with IPV and/or the basal vein. In our data, 74.8% of all cases were of the B
7a type, and the frequency of this type was significantly higher than in previous reports (Yamashita 53.7%;
p < 0.001, Boyden [
2] 22%;
p < 0.001). On the other hand, the B
7ab type and the B
7b type were observed at rates of 14.8 and 4.8%, respectively. The frequencies of these types were significantly lower than those reported by Yamashita (28.5%;
p = 0.001, 10.0%;
p = 0.049). Our results for B
7 differ from those of Yamashita mainly because they defined B
7b and B* as equal, whereas we clearly distinguished between B
7b and B*. Both B* and B
7b bifurcate from the basal bronchus from similar sites; however, B* points in the posterolateral direction directly toward the vertebral body, whereas B
7b points in the mediobasal direction. Furthermore, B
7b mainly crossed over the basal pulmonary vein in our study, unlike in previous reports. The advantage of our study using 3DCTAB was that we could analyze the relationships between vessels and bronchi with the lung parenchyma in its natural inflated state. In contrast, it is sometimes difficult to analyze such anatomical relationships by conventional anatomical techniques using resected lung specimens usually in the deflated state.
We also identified some uncommon drainage patterns of veins, such as aberrant V
4+5, aberrant V
4 or V
5, and aberrant V
6. In particular, aberrant V
6 was not identified in previous reports. These minor anatomical variations in pulmonary veins can cause serious problems in patients undergoing lung surgery [
7,
8]. For example, during right lower lobectomy in a patient with aberrant V
4+5, if we cut an IPV with an aberrant V
4+5, the right middle lobe would become dysfunctional after surgery. Furthermore, during subcarinal lymph node dissection in a patient with aberrant V
6, which runs through the subcarinal area, we may encounter unexpected bleeding if we do not have preoperative knowledge of this uncommon variation. Therefore, knowledge regarding these minor uncommon vessels is necessary to safely perform lung resection.
There were several limitations in this study. First, we could not obtain adequate images of segmental vessels and bronchi from some patients. Although the number of such cases was small, they may have biased our results. Second, our study was an anatomical analysis based on 3DCT findings; therefore, it is possible that there were differences in our data from actual anatomy. The detection rate of vessels by our 3DCTAB was 98.7% [
4], so it appeared feasible to compare our 3DCTAB data with those described in previous reports based on actual anatomy. However, the limitation of differences from actual anatomy remains, because we did not compare our data to resected versions of the same lungs.
Acknowledgements
The authors are grateful to Yasuhiro Fukushima, Junya Fukuda, and Hiroyuki Takei, Department of Radiology, Gunma University Hospital, for obtaining the 3DCTAB images.