Lung cancer (LC) is a common malignancy worldwide and is regarded as the leading cause of death (1). The high mortality of LC puts a tremendous strain on public health systems. In recent years, as high-resolution computed tomography (HRCT) is generally applied to health examinations, ground-glass opacity (GGO) is increasingly being confirmed. Surgery remains the best treatment option for early-stage LC. Studies showed that sublobar resection with adequate surgical margins is feasible and effective (2–6), particularly for lesions with a diameter of less than 2 cm and with a consolidation tumor ratio of less than 25%. The main forms of sublobar resection currently include wedge resection and anatomical segmentectomy. Additionally, anatomical segmentectomy preserves better lung function and minimizes lung volume loss. These benefits draw our attention to video-assisted thoracoscopic surgery (VATS) anatomical segmentectomy. However, the presence of anatomical variations may bring great difficulties and challenges to VATS anatomical segmentectomy.

The procedure of performing VATS anatomical segmentectomy is extremely risky with regard to the pulmonary artery as variable pulmonary artery branches are often encountered during dissociation. Without sufficient preoperative anatomy knowledge, it is difficult to accurately mutilate the artery of the target lung segment intraoperatively, which is likely to lead to the conversion of anatomical segmentectomy to lobectomy, which prolongs the surgical time and leads to loss of lung function. Therefore, a comprehensive understanding of the branching patterns of the peripheral segmental arteries is essential for the successful performance of segmentectomy and important to avoid intraoperative pulmonary vessel injury.

In the early days of lung segmental anatomy research, gross anatomical specimens were the primary source of information (7). In recent years, the technology of three-dimensional computed tomography bronchography and angiography (3D-CTBA) is developing rapidly, which extracts high-quality planar image data from computed tomography (CT) scans and creates three-dimensional (3D) virtual models of the lungs, including segments, subsegments, lesions, bronchi, and vessels. Several studies have analyzed the branching patterns of peripheral segmental arteries in the right upper lobe (RUL). However, there is no report showing the classification of the branching patterns of the interlobar artery and the artery crossing intersegmental planes. The aim of this study was to explore the branching patterns of the interlobar artery and the artery crossing intersegmental planes utilizing data from 3D-CTBA. To further advance our understanding of the branching patterns of the pulmonary artery in the RUL, we also compared the results of our study with a similar study that was previously conducted (8).

With the widespread use of HRCT, increasing numbers of GGO are detected. Some previously published studies have indicated that the prognosis of segmentectomy is no worse than that of lobectomy in patients with early LC (2–6). The anatomical variations of the pulmonary artery make segmentectomy more difficult than lobectomy. Therefore, surgeons must have a comprehensive and accurate understanding of the anatomical characteristics of the branching pattern of the peripheral segmental arteries. Fortunately, advances in the volume-rendering reconstruction technique have enabled the reconstruction of 3D images. 3D-CTBA is a useful tool for thoracic surgeons to identify pulmonary anatomy. An accurate preoperative study can reduce the risk of unexpected bleeding to 2.6% and up to 0 when a printed 3D model is available (10, 11).

However, the branching patterns of the interlobar artery and the artery crossing intersegmental planes are rarely mentioned in previously published reports (7, 8, 12, 13). In the present study, we have comprehensively summarized and classified the branching patterns of the interlobar artery and the artery crossing intersegmental planes using 3D-CTBA.

In the present study ( Table 1 ), the “Tr. sup type” was seen in 25 cases (4.2%), which was lower than that reported by Nagashima (9.9%). The rare anatomic variant of the RUL artery was “Tr. sup + Tr. inf type” (2.5%), which was similar to the findings of Nagashima (3.4%). According to the supplying range and the number of the interlobar artery branch, “Tr. sup + interlobar artery” and “Tr. sup + Tr. inf + interlobar artery” can be respectively divided into four subtypes, which have not been reported in previous literature ( Figure 2 , Table 1 ).

As shown in Figure 2 and Table 1 , the branching patterns of the RUL artery in this study were somewhat different from those in previous reports (7, 8, 12, 13). The main reason is the specific classification of the interlobar artery as described above. Nagashima defined the interlobar artery as the A. asc. However, the interlobar artery can supply S², S³, or S² and S³. In this paper, the interlobar artery of the RUL was classified and summarized in detail by 3D-CTBA. The interlobar artery branch was defined as the A. pos when it only supplied S², while the interlobar artery was defined as the A. ant when it only supplied S³. Similarly, the interlobar artery branch was nominated as the A. asc if it supplied both S² and S³.

Moreover, the definitions of the Tr. inf were ambiguous in the previous studies, so these were redefined in this work. Boyden defined the Tr. inf as the inferior branch of the Tr. sup when it splits into two parts (7). Nagashima defined the Tr. inf as the second branch of the right main pulmonary artery, which arises from the mediastinal portion of the RUL artery between the distal region of the Tr. sup and the proximal region of the first middle lobe of the pulmonary artery (8). However, the two definitions of the Tr. inf are verified in clinical practice ( Figures 1 , 2 ). Furthermore, the boundary between the Tr. inf and interlobar artery branch is controversial. Nagashima defined the boundary as the first middle lobe artery (8), while Boyden defined it as B³ (7). Boyden indicated that the interlobar artery to the RUL occurs below the level of the highest middle lobe artery at 12% and at the level of 40% in the study of bronchovascular variations of the RML (14). Based on 3D-CTBA data from our study, we verified that the originating location of interlobar artery branches emerges at a higher position than the first middle lobe artery in 69.3% (388/560), below the level in 4.5% (25/560), at the level in 26.3% (147/560). We found that the Tr. inf is located at the anterior side of B³ in 97.8% (126/129). Thus, it seems more appropriate to define the boundary as B³.

An understanding of the origin of the interlobar artery branch is significant in clinical practice if a safe and accurate lobectomy is to be performed. The anatomical variation, whereby A⁶ shared a common trunk with the A. pos, was found in 36 patients ( Table 3 ). A⁶ should be carefully separated from the A. pos before it is cut to avoid damaging A⁶ in the RUL lobectomy ( Figure 4A ). Similarly, the A. pos should be carefully separated from A⁶ before it is divided to avoid injuring the A. pos in the right lower lobe (RLL) lobectomy. Additionally, we also found that the A. asc shared a common trunk with A⁶ in seven cases ( Figure 4B , Table 5 ). This variation type is rare in clinical practice; however, if it occurs, it causes a great challenge in the performance of lobectomy. Furthermore, a common trunk for the A. ant and RML artery was observed in 8.2% of the cases ( Figure 4B , Table 4 ). During the RML lobectomy, the A. ant must be protected because a pulmonary artery branch supplied the RUL obstruction that leads to surgical complications such as severe lung edema or extension of the planned lung resection.

The supplying range of the interlobar artery branch has its clinical significance for segmentectomy ( Tables 3– 5). For example, for an accurate S²b segmentectomy, the intersegmental plane is easily altered without the knowledge of the supply range of the A. pos. If the A. pos supplies S²a, a mistaken cut of it will result in an enlarged intersegmental plane ( Figure 4A ). Similarly, if the A. ant supplies S³a, excess cutting of the A. ant will result in an enlarged intersegmental plane in S³b segmentectomy ( Figure 4B ). If the A. asc supplies a part of S³, we need to protect the artery branch supplying S² in the S³ segmentectomy ( Figure 4B ). When the bifurcation of the A. asc is V-shaped after a distance from the root, it is necessary to dissect the A. asc in a center-to-periphery direction to identify the bifurcation in S² or S³ segmentectomy ( Figure 4B ).

The branching patterns noted in the RUL segmental arteries differed greatly from those of previous reports ( Tables 6 – 8 ) (7, 8, 12, 13). This can be explained by the specific classification of the interlobar artery and the artery crossing intersegmental planes. In anatomical segmentectomy, it is significant to understand the branching patterns of pulmonary segmental and subsegmental arteries.

For an accurate S¹ segmentectomy, it is significant to understand the branching patterns of A¹ pre-operatively. We found that the branching patterns of A¹ also had the following branching types: Tr. sup (98.2%), whose incidence is similar to that of Nagashima (98.9%); Tr. sup and Tr. inf (1.8%), which has not been reported in the literature ( Table 6 ). When A¹ branched from the upper segments of the bifurcated Tr. sup or the trifurcated Tr. sup, the Tr. sup must be dissected in a center-to-periphery direction to identify A¹ ( Figures 3 , 5 ). When A¹bii branching from the Tr. inf runs deep within the lung parenchyma, A¹bii can be identified after resection of the B¹ ( Figure 5 ). Additionally, we found that AX¹b bifurcated from A³ in 102 cases ( Figure 5 ). A mistaken cut at the upper segments of the bifurcated Tr. sup (A¹a) will result in a narrowed intersegmental plane. Thus, an understanding of the branches and direction of AX¹b before surgery allows the surgeon to carefully peel off this branch during the intraoperative anatomy, avoiding injury to A³.

The branching patterns of A² have significant clinical significance for accurate S² segmentectomy ( Table 7 , Figure 6 ). A² can be composed of the following five components: A. pos, A. asc, A. rec, AX. rec, and Tr. inf ( Figures 1 , 3 , 6 ) (9). Moreover, the AX. rec is also a new concept (9). Therefore, three basic approaches to identifying these branches were first reported in our previous paper (9). The A. pos and A. asc can be discriminated by dissecting interlobar fissures (interlobar approach). The Tr. sup can be dissected in a center-to-periphery direction to recognize the A. rec and AX. rec (center-to-periphery approach). The Tr. inf running deep within the lung parenchyma and supplying S² was distinguished after resection of B² (posterobronchial approach). If the A. pos supplies S² ( Figure 6 ), the A. pos can be identified by adopting the interlobar approach. If the A. pos and AX. rec supply S², we need to use the interlobar approach and center-to-periphery approach to recognize the A. pos and AX. rec. If A² branched from the A. pos, A. rec, and Tr. inf, the interlobar approach, center-to-periphery approach, and posterobronchial approach should be respectively applied to identify the A. pos, A. rec, and Tr. inf.

Likewise, for an accurate S³ segmentectomy, we need to comprehend that S³ is supplied by how many arterial branches ( Table 8 , Figure 7 ). If the Tr. sup supplies S³, the Tr. sup can be dissected in a center-to-periphery direction to distinguish A³. Moreover, when the AX. rec or AX¹b share a common trunk with A³ ( Figure 3 ), it is necessary to fully dissociate along the Tr. sup intraoperatively to facilitate the disconnection of A³ and the protection of AX. rec, or AX¹b. When the Tr. inf supplies S³, the Tr. inf can be distinguished by dissecting the mediastinal portion of the right main pulmonary artery. If the A. ant supplies S³a and the Tr. sup supplies A³b ( Figure 7 ), the A³a can be identified by dissecting interlobar fissures, and the Tr. sup should be dissected in a center-to-periphery direction to identify A³b. Therefore, it is significant to conduct a comprehensive and thoughtful investigation of anatomical variations preoperatively.

The incorrect vascular identification may lead to surgical complications in segmentectomy. Surgical procedure changes according to vascular variations, and therefore, accurate preoperative recognition of variations is a mainstay when planning RUL segmentectomy. Therefore, preoperative 3D-CTBA to understand the branching patterns of the segmental arteries in the RUL is necessary to perform an accurate segmentectomy and subsegmentectomy.

This is the first report to explore the detailed classification of the interlobar artery and the artery crossing intersegmental planes. We believe that our pulmonary artery data and our new nomenclature will facilitate preoperative simulation and intraoperative navigation when RUL segmentectomy is planned and performed.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This retrospective study was approved by the Research Ethics Committee at Hebei General Hospital (No. 2022119). The need for patient consent was waived because of the retrospective nature of the study.

ZL: project design and initiation, data analysis, and manuscript writing. YK: project design and initiation, data analysis, and manuscript writing. BL: project design and initiation, data analysis, and manuscript writing. WL: data collection. XZ: supervisor. All authors contributed to the article and approved the submitted version.