Variations of pulmonary vein drainage critical for lung resection assessed by three-dimensional computed tomography angiography.

BACKGROUND: It is important to understand pulmonary vein drainage pattern variations and their frequency in order to perform safe anatomical pulmonary resection.
METHODS: Variations and frequencies were assessed using three-dimensional computed tomography angiography (3D-CT) in 194 patients. In cases where the tumor or lymph node caused atelectasis or compression of hilar structures, the involved lobes were excluded from the analyses.
RESULTS: We confirmed variant drainage patterns in 15/189 (8.0%) patients in the right upper lobe (RUL), 29/189 (15.3%) in the right middle lobe (RML), 18/192 (9.5%) in the right lower lobe (RLL), and 5/187 (2.6%) in the left upper lobe (LUL). There was no variant type in the left lower lobe (LLL). There were 14 (7.4%) cases of anomalous superior posterior pulmonary vein of RUL (V2 ) drainage: V2 draining to the superior pulmonary vein (SPV) (n = 2, 1.1%), V2 to the inferior pulmonary vein (IPV) (n = 7, 3.7%), V2 to the left atrium (LA) (n = 2, 1.1%), and V6 to the apical pulmonary vein of the RLL (n = 3, 1.6%). There was a posterior pulmonary vein, V3 to RML pulmonary vein in one case (0.5%). The RML pulmonary vein drained into the IPV in 14 (7.4%) and into the LA in 15 (7.9%) cases. The right V6 directly drained into the LA in 15 (7.9%) and V6 into the SPV in 3 (1.6%) cases. The lingular pulmonary vein drained into the IPV in one case (0.5%) and into the LA in two cases (1.1%). The inferior lingular pulmonary vein V5 drained into the IPV and into the LA in one case (0.5%), respectively.
CONCLUSION: We describe anomalous pulmonary venous drainage patterns and their frequencies particular to anatomic surgical resection. 3D-CT is useful to find such variations.

Introduction

It is important to understand pulmonary vein variations and frequency in order to perform anatomical pulmonary resections, such as pneumonectomy, lobectomy, and segmentectomy, safely for lung cancer. The anatomy of pulmonary vessel structures is important for thoracic surgery, especially during video‐assisted thoracic surgery (VATS), because of the limitation of recognizing three‐dimensional structures on a two‐dimensional monitor. Human lung anatomy is studied using surgical specimens, cadavers, and two‐dimensional images such as chest X‐rays and computed tomography (CT).1 Recently, with advances in CT technology, preoperative evaluation of individual anatomy has been reported.2, 3 Hagiwara et al. reported that preoperative 3D‐CT evaluation led to shorter surgical duration.4 They also preoperatively reported anomalous pulmonary veins in five out of 125 cases, with no intraoperative injury to the pulmonary veins. Oizumi et al. described the utilities of preoperative 3D‐CT for VATS pulmonary segmentectomy after completing the procedure in 98% of 51 consecutive cases without mortality.5

Recently, several papers have described variation and frequency of pulmonary artery branching in surgical cases using 3D‐CT.6, 7, 8 Previous studies have also reported variations in pulmonary vein drainage.1 Taking advantage of 3D‐CT, Cronin et al. and Thorning et al. reported pulmonary vein anatomy for abrasion procedures; however, these studies did not address anatomic lung resection.9, 10 In this study, we analyzed the drainage patterns of pulmonary veins to understand the variations and frequencies to plan safe lung resection and avoid pulmonary vein injuries.

Methods

We evaluated the records of 194 patients who underwent anatomic pulmonary resection, including pneumonectomy, lobectomy, bi‐lobectomy, and segmentectomy following preoperative 3D‐CT at Hokkaido University Hospital between February 2007 and September 2010. In cases that the tumor or involved lymph node caused atelectasis or compression of hilar structures, the involved lobes were excluded from the analyses. Finally, 189 right upper lobes (RULs) and right middle lobes (RML), 192 right lower lobes (RLL), 187 left upper lobes (LUL), and 188 left lower lobes (LLL) were reviewed.

Patients underwent a CT (Aquilion 64; Toshiba, Tokyo, Japan) scan after intravenous administration of an iodinated contrast medium (400 mg I/kg) using a mechanical injector (Dual Shot GX; Nemoto Kyorindo, Tokyo, Japan) to their upper limb for 13 seconds of high pressure and normal saline injected for 25 seconds. Radiological technicians constructed 3D‐CT images of the pulmonary vessels and bronchi using four colors (Fig 1a). CT scanning was performed using the following parameters: gantry rotation speed of 0.35 second per rotation, 0.5 mm collimation, 45 helical pitch, tube voltage of 120 kV, and tube current of 300 mA. Axial sections (0.5 mm thickness) were reconstructed at 0.3 mm intervals and transferred to a standalone workstation (Zio Software, Tokyo, Japan).

Figure 1

(a) Three‐dimensional reconstructed computed tomography is shown in four colors: the pulmonary artery is shown in red, the pulmonary vein in blue, the bronchus in green, and a tumor in grey. (b) The apical pulmonary vein of the right upper lobe (V2) drains into the right inferior pulmonary vein (IPV). (c) The medial (V4, white arrow) and the lateral (V5, white arrowhead) middle lobe pulmonary veins directly drain into the left atrium. (d) Intraoperative finding during right lower lobectomy while preserving V4 (white arrow) and V5 (white arrowhead). IPV is encircled with a black silk string. (e) The right apical pulmonary vein of the right lower lobe, V6 (white arrowhead), drains into the superior pulmonary vein. (f) The lower lingular vein, V5 drains into the left IPV.

Two thoracic surgeons independently reviewed each CT image. We recorded each pulmonary vein drainage pattern and classified these into common and variant types. We defined “variant type” as drainage of the segmental or RML pulmonary vein directly into the left atrium (LA), segmental pulmonary veins from the upper lobes or RML into the inferior pulmonary vein (IPV), or segmental pulmonary veins from lower lobes into the superior pulmonary veins (SPV). When there was discrepancy over branching patterns between investigators, the case was again reviewed with 3D‐CT and thin slice axial views, and consensus was reached. Pulmonary segments were defined according to the international nomenclature accepted in 1949.11 The Hokkaido University institutional review board approved this retrospective study.

Results

We confirmed variant drainage patterns of the pulmonary vein in 15/189 (8.0%) RULs, 29/189 (15.3%) RMLs, 18/192 (9.5%) RLLs, and 5/187 (2.6%) LULs. There was no variant type in LLLs. The variant pulmonary vein was much more frequent on the right (32.8%) than the left (2.6%) side.

There were 14 (7.4%) cases of anomalous superior posterior pulmonary vein in the RUL (V2) drainage (Table 1): V2 draining to the SPV (n = 2, 1.1%), V2 to the IPV (n = 7, 3.7%), V2 to the LA (n = 2, 1.1%), and to the apical pulmonary vein of the RLL (V6, n = 3, 1.6%) (Fig 1b). There was a V3 to RML pulmonary vein in one case (0.5%) and a report of anomalous V2 directly.

Table 1

Segmental or middle lobe pulmonary veins with anomalous drainage

Lobes (n)	Pulmonary veins	Drainage points	This study		Yamashita1	Akiba et al.12	Fourdrain et al.13
			Number of variants	%	%	%	%
RUL (189)			15	8.0
	V2	SPV	2	1.1
		IPV	7	3.7	2.4
		V6	3	1.6
		LA	2	1.1
	V3	RMLV	1	0.5	18.2
RML (189)			29	15.3	6.0	13.5	9.0
		IPV	14	7.4	6.0	5.6	5.0
		LA	15	7.9		7.9	4.0
RLL (192)			18	9.5
	V6	SPV	3	1.6
	V6	LA	15	7.9
LUL (187)			5	2.6		2.5	11.0
	Lingular	IPV	1	0.5			1.0
		LA	2	1.1			10.0
	V5	IPV	1	0.5
		LA	1	0.5
LLL (188)			0	0.0		1.7

IPV, Inferior pulmonary vein; LA, left atrium; LLL, left lower lobe; LUL, left upper lobe; RLL, right lower lobe; RML, right middle lobe; SPV, superior pulmonary vein; V2, posterior pulmonary vein of the upper lobe; V3, anterior pulmonary of the upper lobe; V5, inferior lingular pulmonary vein; V6, apical pulmonary vein of the lower lobe.

In RML, the pulmonary vein drained into the IPV in 14 (7.4%) and into the LA in 15 (7.9%) cases as variant types. Figure 1c shows V4 and V5 directly draining into the LA, while Figure 1d shows the intraoperative finding during right lower lobectomy while preserving the pulmonary veins.

In RLL, V6 directly drained into the LA in 15 (7.9%) and V6 into the SPV in 3 (1.6%) cases (Fig 1e).

In the LULs, lingular pulmonary veins drained into the IPV in one (0.5%) and into the LA in two (1.1%) cases. The inferior lingular pulmonary veins (V5) drained into the IPV and into the LA in one case (0.5%), respectively. Figure 1f shows V5 draining into the IPV.

Table 1 combines our results with data from two articles that reported bilateral pulmonary drainage patterns during pulmonary resection,12, 13 and in a textbook by Yamashita. Yamashita's publication is one of the most read and cited textbooks describing the anatomy of pulmonary segments.1

Discussion

In this study, we describe variations of pulmonary vessel drainage patterns assessed by 3D‐CT angiography. We reviewed 192 cases and investigated segmental and middle lobe pulmonary veins. The number of variants described here is comparable to previous reports.1, 12, 13 However, we detail the drainage points, such as the SPV, IPV, RML pulmonary vein, V6, and LA, more clearly than in previous studies. Variant types were more common on the right (32.8%) than the left (2.6%) side.

Anomalous drainage of right V2 occurred in 7.5% of cases. The most frequent site of drainage was the IPV (3.7%), followed by V6 (1.6%), the LA (1.1%), and the SPV (1.1%). Pulmonary veins must be carefully resected at RULs and should be protected during RLL and/or mediastinal lymph node dissection to avoid injury.

Preservation of an anomalous RML pulmonary vein is essential during both RUL and RLL resection. RML pulmonary vein drainage into the IPV or LA occurs in 15.3% of cases. Surgeons should pay particular attention to avoid injury during RLL resection. The frequency of variants detected in this study is comparable to three previous reports (Table 1). Akiba et al. found V4 draining into the IPV on preoperative 3D‐CT and reported that they could avoid injury during RLL resection.14 They also described the preservation of the RML pulmonary vein directly draining into the LV during VATS RLL resection.15

Anomalous drainage of the right V6 occurs in 9.5% of cases. Drainage into the SPV was observed in 1.6% of cases. This rare anomaly was not reported in the three articles we used to compare our results.1, 12, 13

All anomalous pulmonary vein drainages involved the lingular pulmonary veins (2.6%). The inferior branch of the lingular pulmonary vein (V5) is more likely to drain into the LA, followed by the IPV. The frequency of this anomaly in our study is consistent with previous reports.12 These anomalies are significant during LLL resection. Akiba et al. reported successful VATS LUL resection by recognizing anomalous left V5 drainage into the IPV detected by preoperative 3D‐CT.16 Aragaki et al. reported preoperative recognition of the left V2 draining into V6 on 3D‐CT and the successful preservation of the V5 during VATS LLL resection.17 Ishikawa et al. reported left V2 draining into the IPV and performed LUL resection with successful dissection of the V2 during thoracotomy.18 This V2 variant was not found in our study.

There was no variant type in LLL in our series. Akiba et al. also did not report any anomalous IPV among their 121 cases.12 The variant of the left IPV seems to be relatively rare.

We described anomalous pulmonary venous drainage patterns and their frequencies, particularly in regard to anatomic surgical resection. Three‐dimensional CT is useful to find such variations.

Disclosure

No authors report any conflict of interest.