Computed Tomography–Based Anatomical and Morphometric Analysis of the Left and Right Gastric Arteries in a Vietnamese Population

ABSTRACT

Anatomical variations of the left gastric artery (LGA) and right gastric artery (RGA) are clinically relevant for gastric surgery and endovascular interventions. This study aimed to characterize the origin, anatomical distribution and morphometric features of the LGA and RGA using 64‐slice computed tomography (CT). A cross‐sectional descriptive study was conducted on 385 individuals who underwent contrast‐enhanced abdominal CT at Binh Dan Hospital between September 2024 and April 2025. The LGA originated from the celiac trunk in 91.9% of cases and was most frequently located between the lower one‐third of T12 and the upper one‐third of L1 (> 61%). Its mean diameter and length were 2.7 ± 0.3 mm and 30.3 ± 9.3 mm, respectively, with the Type I branching pattern predominating, characterized by branches supplying the oesophagus, gastroesophageal junction and the lesser curvature. The mean distance from the abdominal aorta to the LGA was 19.6 ± 6.1 mm. The RGA was identified in 88.3% of cases and most commonly arose from the proper hepatic artery (72.9%), with a mean diameter of 1.6 ± 0.2 mm. An arterial arcade along the lesser curvature was observed in 81.3% of cases. These findings provide quantitative anatomical data that may support accurate identification of vascular landmarks during gastric surgery and facilitate appropriate catheter selection for endovascular procedures involving the LGA and RGA.

Abbreviations

CPR

curved planar reconstruction

CT

computed tomography

LGA

left gastric artery

MIP

maximum intensity projection

MPR

multiplanar reconstruction

PACS

Picture Archiving and Communication System

RGA

right gastric artery

SD

standard deviation

VR

volume rendering

1. Introduction

Nowadays, laparoscopic surgery has become a routine approach for the management of various gastric diseases, including gastrointestinal stromal tumours, perforated gastric and duodenal ulcers and particularly gastric cancer at different stages of progression. However, in cases of vascular anatomical variations, surgeons may require additional time to control bleeding and, in some situations, conversion to open surgery is necessary (Usui et al. 2005). Therefore, preoperative assessment of the gastric arterial system plays a crucial role in ensuring safe and effective laparoscopic gastrectomy. Preoperative vascular mapping has been shown to significantly reduce the rate of conversion to open surgery. In addition, recent studies on the branches of the arterial arcade along the lesser curvature have highlighted important interventional applications, such as left gastric artery (LGA) embolization for weight reduction in obese patients and retrograde catheterization via the right and left gastric arteries for embolization or chemotherapeutic infusion in the treatment of primary and metastatic liver cancers, while also preventing gastric metastasis (Hafezi‐Nejad et al. 2019; Imai et al. 2019; Yamagami et al. 2002).

Digital subtraction angiography, although considered the ‘gold standard’ for diagnosing and treating vascular diseases, is an invasive technique. With advances in computed tomography (CT) technology, modern multidetector CT scanners equipped with sophisticated software now provide high‐resolution images with excellent temporal and spatial accuracy, enabling detailed visualization and 3‐dimensional reconstruction of gastric vascular anatomy. Traditionally, vascular anatomy has been studied through international data or cadaveric dissection. However, ethnic and ecological differences may limit the applicability of foreign studies to the Vietnamese population, while cadaver‐based studies are constrained by limited specimen availability and may not adequately represent the general population. Although imaging‐based anatomical studies in living subjects are increasingly common worldwide, such research in Vietnam remains limited.

Therefore, this study aimed to characterize the anatomical features of the left and right gastric arteries and their anastomoses along the lesser curvature using 64‐slice CT in a Vietnamese population and to provide clinically relevant data for surgical and interventional planning.

2. Materials and Methods

2.1. Study Design and Data Source

A cross‐sectional study was conducted using a convenience sampling approach. All abdominal CT scans performed at the Department of Diagnostic Imaging, Binh Dan Hospital, between September 2024 and April 2025 were reviewed. Eligible cases were selected based on the following criteria:

2.2. Eligibility Criteria

2.2.1. Inclusion Criteria

• −Vietnamese adults aged ≥ 18 years.
• −Individuals who underwent contrast‐enhanced upper abdominal CT with vascular assessment, including an arterial phase.
• −Availability of complete clinical and imaging data in the hospital's Picture Archiving and Communication System (PACS)

2.2.2. Exclusion Criteria

Cases with conditions that could alter gastric vascular anatomy were excluded, including:

• −Prior gastric surgery or vascular intervention involving the stomach.
• −Vascular pathologies such as stenosis, aneurysm, malformation, dissection, or thrombosis affecting the gastric arteries, hepatic artery, or superior mesenteric artery, as well as liver cirrhosis.
• −Congenital anomalies or intra‐abdominal tumours in the supramesocolic compartment causing distortion or compression of the gastric vasculature.

2.3. Sample Size

The sample size was estimated using a single population proportion approach with a significance level of α = 0.05 (Z = 1.96) and a margin of error of 0.05. The highest reported anatomical variation rate of 49.3% from Yamagami et al. (2002) was adopted as a conservative estimate, yielding a minimum required sample size of 385 cases, which was achieved in the final analysis.

2.4. Variables and Measurements

General characteristics extracted from medical records included age and sex.

Anatomical characteristics and variations of the LGA included origin, vertebral level of origin, diameter, length, presence of branches, branching pattern, distance from the abdominal aorta and diameter of the parent artery.

Anatomical characteristics and variations of the RGA included detection rate, origin, diameter, diameter of the parent artery and presence of an arterial arcade.

2.5. Data Collection Procedure

Data collection was conducted in three sequential steps.

Step 1: A list of cases was established from all contrast‐enhanced abdominal CT scans, including an arterial phase, performed at the Department of Diagnostic Imaging, Binh Dan Hospital. CT examinations were acquired using a 64‐slice multidetector CT scanner (Brilliance; Philips), with intravenous administration of a non‐ionic, water‐soluble iodinated contrast medium (70–100 mL; 1.5–2 mL/kg) at a rate of 3–4 mL/s. Cases meeting the inclusion criteria were identified, and those meeting the exclusion criteria or presenting conditions that could affect gastric vascular anatomy were excluded.

Step 2: For all eligible cases, demographic and clinical data were extracted using a standardized data collection form. Corresponding CT imaging data were retrieved from the hospital's PACS.

Step 3: Imaging data were analyzed using Carestream Vue PACS software. Reconstruction techniques included multiplanar reconstruction (MPR), maximum intensity projection (MIP), curved planar reconstruction (CPR) and volume rendering (VR). The reconstructed images were evaluated to characterize the gastric vasculature and to obtain morphometric measurements in accordance with the study objectives.

2.6. Statistical Analysis

Data were analyzed using SPSS software (version 20.0). Descriptive statistics were used to summarize the study variables. Categorical variables (e.g., sex and arterial origin) were presented as frequencies and percentages, while continuous variables (e.g., age, arterial diameter and length) were expressed as mean ± standard deviation (SD).

For inferential analysis, the independent samples t‐test was used to compare means between two groups, and the chi‐square (χ ²) test was applied to compare proportions. Correlations between continuous variables (e.g., arterial diameter and length) were assessed using Pearson's correlation coefficient. A p‐value < 0.05 was considered statistically significant.

2.7. Ethical Considerations

Given the retrospective nature of the study and the use of existing imaging data, the requirement for informed consent was waived. All data were anonymized prior to analysis to ensure patient confidentiality. The study was conducted in accordance with the principles of the Declaration of Helsinki.

3. Results

A total of 385 cases were included, with 229 males (59.5%) and 156 females (40.5%). The mean age was 50.3 ± 12.8 years, ranging from 18 to 80 years.

3.1. The Left Gastric Artery

The LGA predominantly originated from the celiac artery (91.9%), followed by the abdominal aorta (7.0%). Other origins, including the hepatogastric trunk, splenic artery and left hepatic artery, together accounted for 1.1% of cases.

The distribution of vertebral levels of the LGA origin is shown in Figure 1. The origin of the LGA clustered predominantly between the T12 and L1 vertebral levels, with the highest frequency at the upper one‐third of L1 (22.3%). Overall, this range accounted for more than 61% of cases.

FIGURE 1.

Distribution of vertebral levels of the origin of the LGA.

Branches of the LGA were identified on CT imaging in 77.1% of cases, with three distinct branching patterns observed. Type I, characterized by branches supplying the oesophagus, gastroesophageal junction and the lesser curvature, was the most common pattern (82.8%). Type II included an additional branch coursing within the lesser omentum (omental branch) and was relatively rare (2.7%). Type III involved a branch to the left hepatic lobe and was observed in 14.5% of cases. The branching patterns of the LGA are illustrated in Figure 2.

FIGURE 2.

Branching patterns of the left gastric artery.

The morphometric characteristics of the LGA according to sex are presented in Table 1. Significant sex‐related differences were observed across all measured morphometric parameters of the LGA, with higher values in males than in females for diameter, length, distance from the abdominal aorta, and parent artery diameter (both p < 0.05).

TABLE 1.

Morphometric characteristics of the left gastric artery by sex.

	Overall (n = 385)	Male (n = 229)	Female (n = 156)	p
	Mean ± SD	Mean ± SD	Mean ± SD
Diameter (mm)	2.7 ± 0.3	2.7 ± 0.3	2.6 ± 0.3	< 0.001
Length (mm)	30.3 ± 9.3	31.8 ± 9.3	28.1 ± 8.8	< 0.001
Distance from the abdominal aorta (mm)	19.6 ± 6.1	20.3 ± 6.0	18.8 ± 6.0	0.023
Diameter of the parent artery (mm)	6.7 ± 3.4	7.1 ± 3.7	6.2 ± 2.7	0.009

As shown in Figure 3, a weak positive correlation was observed between the diameter of the LGA and that of its parent artery (r = 0.138, p = 0.007).

FIGURE 3.

Correlation between the diameter of the LGA and that of its parent artery.

3.2. The Right Gastric Artery

The RGA was identified in 340 cases (88.3%). It predominantly originated from the proper hepatic artery (72.9%), followed by the left hepatic artery (21.8%). Less frequent origins included the gastroduodenal artery (2.6%) and right hepatic artery (1.5%), while the common hepatic artery represented the rarest origin (1.2%).

The morphometric characteristics of the RGA according to sex are presented in Table 2. Both the diameter of the right gastric artery and that of its parent artery were significantly greater in males than in females (p < 0.05).

TABLE 2.

Morphometric characteristics of the right gastric artery by sex.

	Overall (n = 340)	Male (n = 209)	Female (n = 131)	p
	Mean ± SD	Mean ± SD	Mean ± SD
Diameter (mm)	1.6 ± 0.2	1.7 ± 0.2	1.6 ± 0.2	0.005
Diameter of the parent artery (mm)	3.7 ± 0.9	3.8 ± 0.9	3.6 ± 1.0	0.036

As shown in Figure 4, a weak but statistically significant positive correlation was observed between the diameter of the RGA and that of its parent artery (r = 0.277, p < 0.001).

FIGURE 4.

Correlation between the diameter of the RGA and that of its parent artery.

In addition, an anastomosis between the left and right gastric arteries was observed in 81.3% of cases. This proportion increased to 92.1% in cases where both arteries were identified.

4. Discussion

4.1. The Left Gastric Artery

CT‐based studies by Iino et al. (2022) and Yuasa et al. (2016) reported that the LGA originated from the celiac trunk in approximately 85%–90% of cases, consistent with our findings. Variations accounted for about 10%–15%, most commonly involving origins from the abdominal aorta or splenic artery. Similarly, cadaveric studies by Saldarriaga et al. (2023) and Cuong (2011) reported comparable patterns, although with lower proportions of celiac trunk origin (60%–75%) and a higher frequency of anatomical variations (25%–40%). This discrepancy may be attributed to the smaller sample sizes in cadaveric studies, which could increase the relative proportion of observed variations.

From an anatomical perspective, knowledge of vertebral level distribution may assist radiologists in differentiating the LGA from the common hepatic and splenic arteries on imaging studies and in identifying anatomical variations. In the present study, the origin of the LGA was most frequently located between the lower one‐third of T12 and the upper one‐third of L1, accounting for more than 61% of cases.

To the best of our knowledge, no CT‐based studies evaluating the diameter of the LGA have been reported, with most available data derived from cadaveric studies. In the present study, the mean diameter was 2.7 ± 0.3 mm, which is comparable to the cadaveric findings reported by Cuong (2011) (2.5 mm), but smaller than that reported by Saldarriaga et al. (2023) (3.5 ± 0.8 mm). These differences may reflect population variation or methodological differences between imaging‐ and dissection‐based studies.

Similarly, the mean length of the LGA in our study (30.3 ± 9.3 mm) was comparable to that reported by Saldarriaga et al. (2023) (28.7 ± 8.8 mm), remaining within a similar range. Minor differences may be attributed to methodological factors, as cadaveric measurements can be affected by vessel contraction, collapse, loss of elasticity, and chemical fixation, whereas CT‐based measurements in living subjects may better reflect in vivo vascular dimensions. Together, these findings suggest that the length of the LGA is relatively consistent across populations.

In addition to morphometric characteristics, branches of the LGA were identified on CT imaging in 77.1% of cases, although comparable data on this proportion remain limited in the literature. The most common pattern was Type I, characterized by branches supplying the distal oesophagus, cardia, and gastric fundus, consistent with previous findings (Cuong 2011). Some authors have described Type III as a branch supplying the left hepatic lobe, with reported proportions ranging from 20% to 25%. In our study, however, this variant was observed at a lower frequency of 14.5%. This represents an important anatomical variation, and its identification is particularly relevant in hepatic surgery to avoid intraoperative bleeding and postoperative hepatic ischemia, especially in cases where it constitutes the sole arterial supply to the left hepatic lobe. Preservation of this Type III branch, also referred to as a replaced left hepatic artery, has been shown to reduce postoperative transaminase elevation and hepatic infarction in patients undergoing laparoscopic gastrectomy, as reported by Waki et al. (2021). In addition, in transarterial chemoembolization (TACE) and other visceral endovascular interventions, recognition of this variant is essential to prevent missed lesions and to avoid unintended compromise of the left hepatic lobe following gastric interventions.

With respect to spatial relationships, the mean distance from the LGA to the abdominal aorta was 19.65 ± 6.06 mm among cases originating from the celiac trunk, which is similar to the CT‐based findings reported by Abduwani et al. (2024) (20.67 ± 5.05 mm). This similarity may be related to the use of comparable imaging modalities and broadly similar population characteristics.

No previous studies have evaluated the relationship between the size of the LGA and that of its parent artery. In our analysis, a statistically significant positive correlation was observed; however, the correlation coefficient (r < 0.2) indicates that the strength of this association is very weak.

Taken together, these anatomical findings may have practical implications for surgical procedures, particularly lymph node dissection at Station 7 along the LGA, by helping to delineate the anatomical boundaries of the dissection area (Nico et al. 2024).

4.2. The Right Gastric Artery

Compared with previous studies, the visualization rate of the RGA in our study was 88.3%, consistent with that reported by Yamagami et al. (2002). However, higher rates (97%–100%) have been reported by Matsuki et al. (2004), who used a higher injection rate (4–5 mL/s) and contrast dose (2–2.5 mL/kg) than those applied in our protocol (3–4 mL/s and 1.5–2 mL/kg). These technical differences may partly explain the lower visualization rate observed in the present study.

The origin of the RGA demonstrates considerable variability. In our study, the most common origin was the proper hepatic artery (72.9%), consistent with the findings of Usui et al. (2005), but higher than those reported by Matsuki et al. (2004) and Abelleyra Lastoria et al. (2023) (approximately 40%–55%). In contrast, this proportion was lower than that reported by Cuong (2011). Other observed origins included the left hepatic artery, gastroduodenal artery, right hepatic artery, and common hepatic artery. This variability indicates that, although a predominant pattern exists, anatomical variations remain relatively frequent. Therefore, pre‐procedural identification of the RGA is important to reduce potential complications. Particular attention should be paid during lymph node dissection at Station 5 and surgical procedures in the suprapancreatic region to avoid vascular injury. In subtotal gastrectomy, failure to ligate the RGA may result in delayed postoperative bleeding. In addition, during lymph node dissection at STATIONs 5 and 8a in patients with gastric cancer, especially those with antral tumours, surgical techniques may need to be adjusted to avoid injury to the left hepatic artery, which provides an important blood supply to the left lobe of the liver. In cases of transcatheter arterial embolization for bleeding ulcers along the lesser curvature, access through the left hepatic artery may be required when the RGA does not originate from the proper hepatic artery, thereby altering the interventional approach and facilitating effective haemostasis.

The mean diameter of the RGA in our study was 1.6 ± 0.2 mm, which is comparable to that reported by Cuong (2011). In contrast, Saldarriaga et al. (2023) reported a considerably larger diameter (3.2 ± 1 mm), which may reflect population differences or methodological variation.

No previous studies evaluating the relationship between the diameter of the RGA and that of its parent artery were identified. In the present study, a weak but statistically significant positive correlation was observed (r = 0.277, p < 0.05), suggesting that this association is limited in strength. Nevertheless, this relationship may be useful in identifying anatomical variations and in selecting appropriate catheters for surgical or endovascular procedures.

An anastomosis between the right and left gastric arteries was observed in 81.3% of cases, increasing to 92.1% when both arteries were clearly visualized. This finding is consistent with standard anatomical references, including those by Cuong (2011), Gray's Anatomy, and the Atlas of Vascular Anatomy (Uflacker 2020), which describe the arterial arcade along the lesser curvature. In addition, Yamagami et al. (2002) reported successful retrograde catheterization of the RGA via the LGA for embolization prior to hepatic arterial infusion chemotherapy, indirectly supporting the presence of this connection. Our findings therefore further reinforce existing anatomical evidence of this vascular arcade. Clinically, identification of this anastomosis is important in gastrectomy and lymph node dissection along the lesser curvature (Station 3), as it helps define the extent of lymphadenectomy and contributes to safer surgical intervention (Nico et al. 2024).

While these findings provide valuable insights, several limitations should be considered. First, the use of a convenience sampling approach at a single center may limit the generalizability of the findings to other populations. Second, although multidetector CT provides high‐resolution imaging, small arterial branches or subtle anatomical variations may not be fully visualized, potentially leading to underestimation of certain features, such as minor branches or anastomoses. Furthermore, the retrospective nature of the study and reliance on existing imaging data limited the availability of clinical and intraoperative correlation. In addition, measurement accuracy may be influenced by factors such as image reconstruction techniques, slice thickness, and contrast timing. Finally, while this study provides quantitative anatomical data, functional or haemodynamic aspects of the gastric arterial system were not assessed.

This study provides detailed anatomical data on the left and right gastric arteries in a Vietnamese population. These findings may assist surgeons in identifying anatomical landmarks during dissection, thereby enabling precise vascular ligation and division in gastric surgery. In addition, they may support the selection of appropriate catheters for endovascular interventions.

Funding

The authors have nothing to report.

Ethics Statement

Ethical approval was obtained from the Research Ethics Committee of Pham Ngoc Thach University of Medicine (Decision No. 1235/TDHYKPNT‐HDDD).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.