AJR ARRS Membership
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ishigami, K.
Right arrow Articles by Honda, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ishigami, K.
Right arrow Articles by Honda, H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.05.1114
AJR 2006; 187:1002-1009
© American Roentgen Ray Society

Original Research

Accessory Left Gastric Artery from Left Hepatic Artery Shown on MDCT and Conventional Angiography: Correlation with CT Hepatic Arteriography

Kousei Ishigami1, Kengo Yoshimitsu1, Hiroyuki Irie1, Tsuyoshi Tajima1, Yoshiki Asayama1, Masakazu Hirakawa1 and Hiroshi Honda1

1 All authors: Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka City, Fukuoka 812-8582, Japan.

Received June 28, 2005; accepted after revision August 22, 2005.

 
Address correspondence to K. Ishigami (ishigamikousei{at}aol.com).


Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
OBJECTIVE. The objective of our study was to evaluate the diagnostic accuracy of MDCT and conventional angiography in the detection of an accessory left gastric artery using CT hepatic arteriography as the standard of reference.

MATERIALS AND METHODS. The study group consisted of 118 patients who underwent MDCT with a triple-phase liver protocol with a slice thickness of 5 mm, conventional angiography, and CT hepatic arteriography. The early-phase images of MDCT and conventional angiography were retrospectively reviewed. The presence or absence of an accessory left gastric artery was evaluated using CT hepatic arteriography as the standard of reference. The sensitivity, specificity, and accuracy of MDCT and conventional angiography were calculated.

RESULTS. CT hepatic arteriography revealed an accessory left gastric artery in 25 (21.2%) of 118 cases, including 15 proximal- and 10 distal-type accessory left gastric arteries. On MDCT, there were seven false-negative cases and one false-positive case. Six of the seven false-negative cases were a proximal-type accessory left gastric artery, and nine of 10 distal-type accessory left gastric arteries were correctly diagnosed using MDCT. The sensitivity, specificity, and accuracy of MDCT were 72.0%, 98.9%, and 93.2%, respectively. On conventional angiography, there were three false-negative and two false-positive cases, none of whom underwent selective left hepatic arteriography. Two of the three false-negative cases were the distal-type accessory left gastric artery, but eight of the 10 distal-type accessory left gastric arteries were correctly diagnosed. The sensitivity, specificity, and accuracy of conventional angiography were 88.0%, 97.8%, and 95.8%, respectively.

CONCLUSION. Approximately 70% of accessory left gastric arteries can be diagnosed at the early phase of MDCT even with a slice thickness of 5 mm.

Keywords: Anatomy • Ct Arteriography • Digital Subtraction Angiography • Gastric Arteries • Hemodynamics • Hepatic Arteries • Mdct Angiography


Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
An accessory left gastric artery arising from the left hepatic artery is a variant gastric artery that supplies the cardia and fundus of the stomach. Like an accessory or replaced left hepatic artery arising from the left gastric artery, an accessory left gastric artery runs through the fissure of the ligamentum venosum, although the blood flow is in the opposite direction [1]. The incidence of this anatomic variant has been reported to be 3-14% in angiographic studies [2, 3]. On angiography, an accessory left gastric artery is seen to arise between the origin of the left hepatic artery and the umbilical point where the left hepatic artery kinks at the umbilical fossa [3].

Hepatic arterial infusion chemotherapy is a treatment of choice for unresectable liver tumors. Percutaneous port-catheter system placement methods using angiographic techniques were developed as a substitute for surgical catheter insertion [4]. During these interventional procedures, the right gastric artery is usually embolized by metallic coils to prevent gastric complications (e.g., ulcer) caused by chemotherapeutic agents. Because the right gastric artery typically arises from the proper hepatic artery, it is not difficult to recognize on angiography. However, an accessory left gastric artery may be mistakenly interpreted to be intrahepatic branches of the left hepatic artery [3, 5-7] and the accuracy of conventional angiography has not been evaluated.

MDCT angiography has been widely used for the evaluation of liver arteries, including the preoperative evaluation of hepatic resections [8, 9], living related liver transplantation donors [10-12], and surgical placement of hepatic arterial infusion chemotherapy pumps [13, 14]. However, little attention has been paid to the diagnosis of an accessory left gastric artery. In addition, MDCT angiography may not be indicated in the setting in which interventional angiographic procedures are subsequently planned. Furthermore, raw MDCT data may not be available in all cases, and retrospective reconstruction with thinner-slice images may not always be possible. Our practical question is, How accurately can MDCT performed using a liver protocol show an accessory left gastric artery in these particular clinical settings?


Figure 1
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1A —77-year-old man with proximal-type accessory left gastric artery. Hepatic arterial phase image of contrast-enhanced MDCT shows vessel running through fissure of ligamentum venosum (white arrow). Note tortuosity of vessel near cardia of stomach. Black arrow indicates origin of accessory left gastric artery.

 


Figure 2
View larger version (116K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1B —77-year-old man with proximal-type accessory left gastric artery. Digital subtraction angiography image from left hepatic artery shows accessory left gastric artery, origin of which is midportion of left hepatic artery (large arrow). Gastric wall stain (arrowhead) is seen. Esophageal branch from accessory left gastric artery (small arrow) is noted.

 


Figure 3
View larger version (141K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1C —77-year-old man with proximal-type accessory left gastric artery. CT hepatic arteriography image from left hepatic artery shows intensely enhancing vessel in fissure of ligamentum venosum (white arrow) with contrast enhancement of cardia and fundus of stomach (black arrow), consistent with an accessory left gastric artery.

 
CT hepatic arteriography combined with conventional angiography and CT during arterioportography is an invasive diagnostic study for the preoperative evaluation of hepatic tumors that has been used for the detection and characterization of hepatocellular carcinoma [15, 16]. On CT hepatic arteriography, extrahepatic branches from the hepatic artery and their vascular territories show contrast enhancement because hepatic arterial contrast material is typically injected from the proper hepatic artery. Therefore, CT hepatic arteriography could depict an accessory left gastric artery, if it exists, more confidently than conventional angiography [7] by showing an enhancing vessel through the fissure of the ligamentum venosum with contrast enhancement of the proximal gastric wall.

The purpose of this study was to assess the diagnostic accuracy of MDCT and conventional angiography for identification of an accessory left gastric artery using CT hepatic arteriography as the standard of reference.


Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patient Population
Between February 2003 and March 2005, 166 consecutive patients underwent conventional angiography, CT hepatic arterioportography, and CT hepatic arteriography at our institution. Because we used CT hepatic arteriography as the standard of reference for diagnosing an accessory left gastric artery, the study coordinator reviewed diagnostic reports of conventional angiography and CT hepatic arteriography to ascertain whether contrast material had been injected from the proper, left, or common hepatic artery. Thirty-six patients were excluded because they had undergone CT hepatic arteriography from the right hepatic artery (n = 32), medial segmental branch of the left hepatic artery (n = 2), cholecystic artery (n = 1), or right inferior phrenic artery (n =1). Of the 32 patients who underwent CT hepatic arteriography from the right hepatic artery, 18 patients had variant right or left hepatic arteries (or both). In two cases, the studies were technically inadequate because the tip of the angiographic catheter migrated into the gastroduodenal artery during CT hepatic arteriography. In addition, seven patients were excluded because they had a history of left hepatic lobectomy (n =6) or total gastrectomy (n = 1). Furthermore, in three cases, MDCT was not available because CT studies were performed at outside hospitals. Therefore, the study group consisted of the remaining 118 patients, all of whom underwent MDCT with a triple-phase liver protocol, conventional angiography, and CT hepatic arteriography at our institution. The institutional review board did not require that we seek approval for this type of retrospective study.


Figure 4
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2A —71-year-old man (patient 10 in Table 2) with distal-type accessory left gastric artery that was not recognized on conventional angiography. Digital subtraction angiography image from proper hepatic artery shows accessory left gastric artery arising near umbilical point (black arrow). In retrospect, gastric wall stain (white arrow) is noted overlying lateral edge of liver. Arrowhead indicates right gastric artery.

 

View this table:
[in this window]
[in a new window]

  		TABLE 2: False-Negative Cases in the Diagnosis of Accessory Left Gastric Artery

 


Figure 5
View larger version (107K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2B —71-year-old man (patient 10 in Table 2) with distal-type accessory left gastric artery that was not recognized on conventional angiography. Vessel within ligamentum venosum (arrow) is seen on hepatic arterial phase image of contrast-enhanced MDCT. This vessel is continuous to left hepatic artery (not shown), suggestive of accessory left gastric artery. Asterisk indicates splenic infarction after partial splenic embolization that occurred previously.

 


Figure 6
View larger version (104K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2C —71-year-old man (patient 10 in Table 2) with distal-type accessory left gastric artery that was not recognized on conventional angiography. CT hepatic arteriography images from proper hepatic artery clearly reveal contrast enhancement of cardia of stomach (arrow, C) and accessory left gastric artery (arrow, D).

 


Figure 7
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2D —71-year-old man (patient 10 in Table 2) with distal-type accessory left gastric artery that was not recognized on conventional angiography. CT hepatic arteriography images from proper hepatic artery clearly reveal contrast enhancement of cardia of stomach (arrow, C) and accessory left gastric artery (arrow, D).

 
The study group was composed of 88 men and 30 women (age range, 23-84 years; mean age, 65.3 years). Typically, the indication of conventional angiography and CT hepatic arteriography was preoperative evaluation of hepatic tumors. Preoperative or tentative diagnoses included hepatocellular carcinoma (n = 104), cholangiocellular carcinoma (n = 8), and metastasis (n = 4). In two cases, conventional angiography and CT hepatic arteriography were performed for further evaluation of questionable liver lesions in patients with idiopathic portal hypertension (n = 1) and portal vein occlusion (n = 1).

Imaging Techniques
Triple-phase CT (i.e., hepatic arterial, portal venous, and delayed phases) was performed with a 4-MDCT scanner (Aquilion, Toshiba Medical Systems; or Somatom Plus 4 Volume Zoom, Siemens Medical Solutions-Asahi Medical Technologies) equipped with a gantry rotation time of 0.5 seconds. Each patient received 100 mL of IV nonionic contrast material containing 370 mg I/mL (iopamidol [Iopamiron, Schering]) or 350 mg I/mL (iohexol [Omnipaque 350, Daiichi Pharmaceutical]) by means of an automated power injector at a rate of 2.5 mL/s. The entire liver was scanned in a craniocaudal direction. With the Aquilion unit, the scanning parameters were as follows: section collimation, 3 mm; section thickness, 5 mm; reconstruction interval, 5 mm; helical pitch, 5.5; 300 mAs; and 120 kVp. With the Somatom Plus 4 Volume Zoom unit, the scanning parameters were as follows: section collimation, 2.5 mm; section thickness, 5 mm; reconstruction interval, 5 mm; table feed, 13.8 mm/gantry rotation; 110 effective mAs; and 120 kVp. Therefore, the parameters were almost equivalent. The hepatic arterial phase images were obtained at a fixed delay of 45 seconds after the start of IV contrast material injection. The portal venous and delayed phase images were acquired at 70 and 240 seconds after the start of contrast material injection, respectively.


Figure 8
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3A —67-year-old man with diaphragmatic branch from accessory left gastric artery. Digital subtraction angiography image from proper hepatic artery shows accessory left gastric artery and esophageal branch (small arrows). Diaphragmatic branch from origin of accessory left gastric artery (large arrows) is noted. However, it was thought to be intrahepatic arterial branch. Vasospasm of proper hepatic artery is seen.

 


Figure 9
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3B —67-year-old man with diaphragmatic branch from accessory left gastric artery. CT hepatic arteriography image reveals arterial branch toward diaphragm (black arrow). Contrast enhancement of esophageal wall (white arrow) is also noted.

 


Figure 10
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4A —Schematic drawing of two types of accessory left gastric artery: proximal (A) and distal (B) types. ALGA = accessory left gastric artery, PHA = proper hepatic artery, RHA = right hepatic artery, LHA = left hepatic artery, A2 = dorsolateral branch, A3 = ventrolateral branch, A4 = medial segmental branch, UP = umbilical point. For proximal type, origin of accessory left gastric artery (arrow) is between origin and proximal two thirds of left hepatic artery.

 


Figure 11
View larger version (17K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4B —Schematic drawing of two types of accessory left gastric artery: proximal (A) and distal (B) types. ALGA = accessory left gastric artery, PHA = proper hepatic artery, RHA = right hepatic artery, LHA = left hepatic artery, A2 = dorsolateral branch, A3 = ventrolateral branch, A4 = medial segmental branch, UP = umbilical point. For distal type, origin of accessory left gastric artery (arrow) is near umbilical point.

 
CT hepatic arteriography was performed as part of a comprehensive angiographic examination combined with conventional angiography and CT during arterioportography. Written informed consent was obtained from all of the patients before angiography. The Seldinger technique was used to place two sets of 4-French sheath-catheter systems, typically through the right femoral artery. Hepatic arterial anatomy, portal venous patency, and the presence of hepatic tumor were assessed with digital subtraction angiography before CT during arterioportography and CT hepatic arteriography. On conventional angiography, selective left hepatic arteriography was performed in 68 of 118 patients. The hepatic arterial catheter for CT hepatic arteriography was placed in the proper hepatic artery (n = 86), left hepatic artery (n = 21), or common hepatic artery (n = 11). CT hepatic arteriography was performed from the right and left hepatic arteries separately in 13 patients, nine of whom had variant right or left hepatic arteries (or both). CT hepatic arteriography was performed approximately 5 min after CT during arterioportography. Dual-phase CT hepatic arteriography was performed using a 4-MDCT scanner (Aquilion) in a craniocaudal direction during single breath-hold acquisitions of 25-35 seconds, depending on liver size. Scanning parameters for CT hepatic arteriography were identical to those for triple-phase MDCT. Data acquisitions started 15 seconds after initiation of a transcatheter arterial injection of 16-50 mL of non-ionic contrast material (iopamidol 150 mg I/mL [Iopamiron 150, Schering]) at a rate of 0.8-2.5 mL/s (20 seconds of injection duration) using an automated power injector. The approximate injection rate for CT hepatic arteriography was determined as the maximum injection rate that would not cause backward flow of contrast material on digital subtraction angiography.

Imaging Analysis
MDCT, conventional angiography, and CT hepatic arteriography data were transferred to a commercially available computer workstation (TWS-2000, Toshiba Medical Systems) for review of hepatic arterial anatomy.

Axial images of the hepatic arterial phase of MDCT studies were retrospectively reviewed by two of the authors without knowledge of the findings of conventional angiography and CT hepatic arteriography. The presence or absence of an accessory left gastric artery was assessed by consensus of the two reviewers. On MDCT, an accessory left gastric artery was shown to be a vessel running through the fissure of the ligamentum venosum (Figs. 1A and 2B). Because an accessory or replaced left hepatic artery arising from the left gastric artery also runs through the fissure of the ligamentum venosum, the reviewers carefully followed the course of the left hepatic and gastric arteries; an accessory left gastric artery is separate from the left gastric artery, and vice versa. An accessory left gastric artery is continuous to the left hepatic artery and appears tapered or tortuous near the cardia of the stomach, at the distal portion of the accessory left gastric artery. In contrast, an accessory or replaced left hepatic artery is continuous to the left gastric artery and appears smaller as it goes through the intrahepatic portion. Reviewers also evaluated variant hepatic arterial anatomy (e.g., replaced right hepatic artery). Neither multiplanar reformation nor maximum intensity projection was used for reviewing the MDCT studies.

Conventional angiography studies were retrospectively reviewed by two of the authors. The reviewers assessed the studies for the presence of an accessory left gastric artery and variant hepatic arterial anatomy by consensus. To determine whether an accessory left gastric artery existed, the reviewers used selective angiography such as left (n = 68), proper (n = 43), and common (n = 7) hepatic arteriography. Celiac and superior mesenteric arteriography images were used to assess variant hepatic arteries. In addition, if reviewers noticed other extrahepatic branches arising from an accessory left gastric artery (Figs. 1B and 3A), they were recorded.

The findings of CT hepatic arteriography were used as the standard of reference. CT hepatic arteriography studies were reviewed by the study coordinator. On CT hepatic arteriography, an accessory left gastric artery was depicted as an enhancing vessel within the fissure of the ligamentum venosum (Figs. 1C and 2D). Contrast enhancement of the cardia and fundus of the stomach was seen (Figs. 1C and 2C). If extrahepatic branches from an accessory left gastric artery existed, CT hepatic arteriography also showed enhancing vessels with contrast enhancement of the corresponding vascular territories (Fig. 3B).

To further evaluate false-negative cases, we divided the origin of the missed accessory left gastric artery into two types: proximal and distal (Fig. 4A and 4B). This classification was modified from the study by Shioyama et al. [7]. The proximal type was defined as occurring when the origin of accessory left gastric artery was between the origin and the proximal two thirds of the left hepatic artery (Figs. 1A, 1B, 1C, 3A, 3B, and 4A). The distal type was defined as occurring when the origin was near the umbilical point (Figs. 2A, 2B, 2C, 2D, and 4B).

Data Analysis
The sensitivity, specificity, and accuracy in the diagnosis of an accessory left gastric artery on MDCT and conventional angiography were calculated using the findings of CT hepatic arteriography as the standard of reference. Conventional angiographic findings of extrahepatic branches from accessory left gastric artery were compared with those of CT hepatic arteriography.


Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
CT hepatic arteriography revealed an accessory left gastric artery in 25 (21.2%) of the 118 patients, including 15 proximal-type (60.0%) and 10 distal-type (40%) accessory left gastric arteries. Extrahepatic arterial branches from the accessory left gastric artery were noted in 24 (96.0%) of the 25 cases. Table 1 summarizes the data for cases of extrahepatic branches from an accessory left gastric artery.


View this table:
[in this window]
[in a new window]

  		TABLE 1: Extrahepatic Arterial Branches from an Accessory Left Gastric Artery

 

On MDCT, an accessory left gastric artery was correctly diagnosed in 17 cases. There was one false-positive case. This case turned out to be an accessory left hepatic artery from the left gastric artery supplying the dorsolateral subsegment of the left hepatic lobe. There were seven false-negative cases in which MDCT could not reveal the accessory left gastric artery. The sensitivity, specificity, and accuracy of MDCT were 72.0%, 98.9%, and 93.2%, respectively. Six (85.7%) of seven false-negative MDCT cases were proximal-type accessory left gastric artery. Nine (90%) of 10 distal-type accessory left gastric artery cases were correctly diagnosed on MDCT. In contrast, only nine (60.0%) of 15 proximal-type accessory left gastric artery cases could be recognized. However, this was not statistically significant (p = 0.102, chi-square test). The cases with false-negative MDCT or false-negative conventional angiography findings are summarized in Table 2.

Conventional angiography correctly showed an accessory left gastric artery in 22 patients. There were two false-positive cases. In one case, the right gastric artery from the left hepatic artery was mistaken for an accessory left gastric artery (Figs. 5A and 5B). In the remaining case, an intrahepatic branch of the left hepatic artery was mistaken for an accessory left gastric artery. There were three false-negative cases in which an accessory left gastric artery was mistakenly considered to be an intrahepatic branch of the left hepatic artery (Figs. 2A, 2B, 2C, and 2D). In these five false-positive and false-negative cases, selective left hepatic arteriography was not performed. The sensitivity, specificity, and accuracy of conventional angiography in diagnosing an accessory left gastric artery were 88.0%, 97.8%, and 95.8%, respectively. In three cases with false-negative conventional angiography findings, two (66.7%) were the distal type. In the remaining case, the peripheral portion of the accessory left gastric artery was not included in the angiography films. Therefore, 14 (93.3%) of 15 proximal-type and eight (80%) of 10 distal-type accessory left gastric arteries were correctly diagnosed using conventional angiography. In addition, extrahepatic branches from an accessory left gastric artery were recognized on conventional angiography in 18 patients (Table 1), including the lower esophagus, diaphragm, mediastinum, and pericardium.


Figure 12
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5A —70-year-old man with right gastric artery arising from left hepatic artery, which was thought to be accessory left gastric artery. Digital subtraction angiography image from common hepatic artery shows artery arising from proximal left hepatic artery (arrow), mimicking accessory left gastric artery. Tumor stain in lateral segment of left hepatic lobe (arrowhead) is also noted, representing hepatocellular carcinoma.

 

Figure 13
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5B —70-year-old man with right gastric artery arising from left hepatic artery, which was thought to be accessory left gastric artery. CT hepatic arteriography from proper hepatic artery shows contrast enhancement of lesser curvature of distal stomach (white arrow) corresponding to vascular territory of right gastric artery (black arrow).

 
Conventional angiography revealed 18 (15.2%) of 118 patients with variant hepatic artery, including an accessory or replaced left hepatic artery (n = 7), a replaced right hepatic artery (n = 5), an accessory or replaced left hepatic artery and replaced right hepatic artery (n = 4), and hepatomesenteric trunk (common hepatic artery from the superior mesenteric artery, n = 2). MDCT correctly characterized 16 (88.8%) of 18 of patients with variant hepatic artery. However, in three cases, hepatic arterial anatomy was incorrectly characterized on the basis of MDCT. In the first case, an accessory left hepatic artery was thought to be an accessory left gastric artery. In the second case, a replaced right hepatic artery was mistakenly considered to be a hepatomesenteric trunk. In addition, in the third case, standard hepatic arterial anatomy was overcalled as a replaced right hepatic artery. In that case, celiac stenosis and an arterial collateral from the superior mesenteric artery were noted on conventional angiography.


Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The incidence of an accessory left gastric artery in our study group was higher (21.2%) than those reported in previous studies. It has been reported that an accessory left gastric artery is more frequently found in Japanese individuals (11.1%) than in Europeans (2.0%) in autopsy series [17]. In angiographic studies, similar results have been reported. Nakamura et al. [3] reported that the incidence of an accessory left gastric artery was 14.2%. In contrast, Michels [2] reported it was only 3.0% of incidence.

The most important clinical situation in the diagnosis of an accessory left gastric artery is hepatic arterial infusion chemotherapy, transcatheter arterial chemoembolization, or radioembolization. In fact, a case of gastric perforation that developed after hepatic arterial infusion chemotherapy has been reported [6]. If such a condition is recognized, coil embolization of the accessory left gastric artery would be indicated before placement of a port-catheter system. When transcatheter arterial chemoembolization is performed from the left hepatic artery, the tip of the catheter should be advanced distal to the origin of the accessory left gastric artery. In addition, an accessory left gastric artery can be the site for arterial bleeding from the distal esophagus and proximal stomach [3]. Furthermore, gastric cancer fed by an accessory left gastric artery has been reported [18, 19].

To the best of our knowledge, only one study has described the visualization of an accessory left gastric artery on CT. Ohgiya et al. [5] used single-detector helical CT with 10-mm collimation. They found an accessory left gastric artery in only three (1.8%) of 166 patients, and their findings were 100% accurate. This incidence is much lower than that in our study. Ohgiya et al. used conventional angiography as the standard of reference. However, in their study, selective arteriography of the common or proper hepatic artery was performed in 48 of 166 patients. As they admitted, a limitation of celiac arteriography is that there might have been a substantial number of accessory left gastric arteries that were not recognized on either CT or celiac arteriography in their study.

There are several problems with using conventional angiography to diagnose an accessory left gastric artery. The lateral edge of the left hepatic lobe, the spleen, and the stomach may overlap on celiac arteriography. Especially when the liver lesion is noted in the right lobe, the peripheral portion of an accessory left gastric artery may not be included in the angiography films, and even if it is included, an angiographer may not pay enough attention to the presence of an accessory left gastric artery. Misregistration artifacts from digital subtraction angiography may cause a diagnostic problem when evaluating a small gastric wall stain (Figs. 2A, 2B, 2C, and 2D). In addition, when the stomach is contracted, the gastric wall stain from an accessory left gastric artery may mimic a hypervascular mass [3]. Furthermore, Shioyama et al. [7] reported that the distal type of accessory left gastric artery might be difficult to recognize because it runs close to the dorsolateral branch of the left hepatic artery on angiography.

In our study, selective proper or common hepatic arteriography was performed in all 118 patients. However, there were three false-negative and two false-positive cases on conventional angiography, none of whom underwent left hepatic arteriography. Two of the three false-negative cases were the distal type of accessory left gastric artery. One of the false-positive cases was the right gastric artery arising from the left hepatic artery. This type of arterial branching was noted in six (5.1%) of 118 patients in our study group. However, it is not important to distinguish this type of branching from an accessory left gastric artery because the clinical significance is similar in either case. To correctly diagnose an accessory left gastric artery, it may be desirable to perform left hepatic arteriography, and each of the subsegmental arterial branches of the left hepatic artery should be carefully followed.

Nakamura et al. [3] reported that the incidence of an esophageal branch from an accessory left gastric artery was 71.4% in their angiographic study. In our series, it was 92.0% (23 of 25 patients). In addition, we observed various extrahepatic branches from an accessory left gastric artery, including the diaphragm, mediastinum, and pericardium (Table 1). These arterial branches were not previously described in detail and may also cause adverse effects for hepatic arterial infusion chemotherapy or transcatheter arterial chemoembolization. For example, it has been reported that arterial infusion chemotherapy from the inferior phrenic artery was frequently associated with pulmonary complications [20].

The fissure of the ligamentum venosum is easily recognized on axial CT images because it runs a nearly horizontal course. The vessels running through the ligamentum venosum include an accessory or replaced left hepatic artery arising from the left gastric artery, an accessory left gastric artery, and an aberrant left gastric vein. The incidence of an accessory or replaced left hepatic artery has been reported to be 14.5% [21]. Therefore, the incidences of an accessory or replaced left hepatic artery and an accessory left gastric artery are similar. Although there was one misdiagnosed case in our study, on careful review of the left hepatic and left gastric arteries on axial MDCT images, we could distinguish an accessory left gastric artery from an accessory or replaced left hepatic artery. On the other hand, an aberrant left gastric vein is rare, and the incidence has been reported to be 0.06-0.8% [22, 23]. In our study, an aberrant gastric vein was not encountered. However, an aberrant left gastric vein would have been more confidently excluded on MDCT if we had used the earlier scan delay (e.g., 25-30 seconds) because gastric venous return would happen at the scan delay of 45 seconds [24].

MDCT visualization of an accessory left gastric artery in our cases was not highly sensitive (sensitivity of 72.0%). MDCT was more sensitive in depicting the distal type of an accessory left gastric artery (90%) than the proximal type (60%), although the difference was not statistically significant. The distal type runs a relatively horizontal course, and the proximal type runs obliquely [3, 7], which might be a possible explanation for the different levels of sensitivity on the axial MDCT images. Because the proximal type is not difficult to recognize on conventional angiography, it may be beneficial to be aware of the distal type of accessory left gastric artery by reviewing axial MDCT images before interventional angiographic procedures.

There were several limitations to our study. We used axial MDCT images with a low injection rate of 2.5 mL/s and a fixed scan delay of 45 seconds. MDCT angiography with a narrower collimation, higher injection rate, and earlier scan delay is more suitable for the evaluation of a small artery. In our cases, MDCT with a triple-phase liver protocol was performed with a slice thickness of 5 mm because it was reported that there was little or no advantage in reducing slice thickness to less than 5 mm to detect hypervascular hepatocellular carcinoma [25]. MDCT angiography with thin-slice images (isotropic voxel), obtained by the routine use of 16- and 64-slice scanners, would have improved the accuracy of diagnosing an accessory left gastric artery by making use of multiplanar reformation and maximum intensity projection. Further study is necessary to clarify the value of MDCT angiography for detecting an accessory left gastric artery.

We used CT hepatic arteriography as the standard of reference because we attempted to evaluate the diagnostic accuracy of conventional angiography. On CT hepatic arteriography, it might be difficult to distinguish an accessory or replaced left hepatic artery from an accessory left gastric artery when contrast material flows back into the stomach. On conventional angiography, however, it is not a diagnostic problem to diagnose an accessory or replaced left hepatic artery. In contrast, on MDCT, both an accessory or replaced left hepatic artery and an accessory left gastric artery are seen within the fissure of the ligamentum venosum. Because we also would like to evaluate whether MDCT can distinguish an accessory left gastric artery from an accessory or replaced left hepatic artery, we included patients with an accessory or replaced left hepatic artery who underwent CT hepatic arteriography in the study group. In addition, we excluded patients who underwent CT hepatic arteriography from the right hepatic artery. A substantial number of patients (n = 18) who had variant hepatic arteries were excluded. Furthermore, most of our patients had liver cirrhosis or chronic hepatitis. The morphologic changes of the liver in either setting might change the appearance of an accessory left gastric artery on axial CT images. Finally, retrospective reviews by consensus might cause a source of bias in interpretations.

In summary, approximately 70% of the cases of an accessory left gastric artery can be detected on MDCT even with a slice thickness of 5 mm. MDCT is more accurate for diagnosing the distal type of accessory left gastric artery than the proximal type.


References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 1996;9 : 143-150[CrossRef][Medline]
  2. Michels NA. Blood supply and anatomy of the upper abdominal organs with a descriptive atlas. Philadelphia, PA: Lippincott, 1955
  3. Nakamura H, Uchida H, Kuroda C, et al. Accessory left gastric artery arising from left hepatic artery: angiographic study. AJR 1980; 134:529 -532[Abstract]
  4. Tajima T, Yoshimitsu K, Kuroiwa T, et al. Percutaneous femoral catheter placement for long-term chemotherapy infusions: preliminary technical results. AJR 2005;184 : 906-914[Abstract/Free Full Text]
  5. Ohgiya Y, Gokan T, Munechika H. Demonstration of aberrant hepatic and gastric arteries with helical CT. Invest Radiol1999; 34:579 -584[CrossRef][Medline]
  6. Takeuchi N, Shioyama Y. Case of gastric perforation after TAI (transcatheter arterial infusion) of SMANCS with special reference to accessory left gastric artery [in Japanese]. Gan To Kagaku Ryoho 2005; 32:547 -551[Medline]
  7. Shioyama Y, Takasaka I, Onaya H, et al. Topography of the accessory left gastric artery (ALGA) analyzed by CT angiography from the left hepatic artery [in Japanese]. Jpn J Intervent Radiol2003; 18:267 -270
  8. Sahani D, Saini S, Pena C, et al. Using multidetector CT for preoperative vascular evaluation of liver neoplasms: technique and results. AJR 2002; 179:53 -59[Abstract/Free Full Text]
  9. Stemmler BJ, Paulson EK, Thornton FJ, Winters SR, Nelson RC, Clary BM. Dual-phase 3D MDCT angiography for evaluation of the liver before hepatic resection. AJR 2004;183 : 1551-1557[Abstract/Free Full Text]
  10. Lee SS, Kim TK, Byun JH, et al. Hepatic arteries in potential donors for living related liver transplantation: evaluation with multi-detector row CT angiography. Radiology2003; 227:391 -399[Abstract/Free Full Text]
  11. Guiney MJ, Kruskal JB, Sosna J, Hanto DW, Goldberg SN, Raptopoulos V. Multi-detector row CT of relevant vascular anatomy of the surgical plane in split-liver transplantation. Radiology2003; 229:401 -407[Abstract/Free Full Text]
  12. Kamel IR, Kruskal JB, Pomfret EA, Keogan MT, Warmbrand G, Raptopoulos V. Impact of multidetector CT on donor selection and surgical planning before living adult right lobe liver transplantation. AJR 2001; 176:193 -200[Abstract/Free Full Text]
  13. Kapoor V, Brancatelli G, Federle MP, Katyal S, Marsh JW, Geller DA. Multidetector CT arteriography with volumetric three-dimensional rendering to evaluate patients with metastatic colorectal disease for placement of a floxuridine infusion pump. AJR 2003;181 : 455-463[Abstract/Free Full Text]
  14. Sahani DV, Krishnamurthy SK, Kalva S, et al. Multidetector-row computed tomography angiography for planning intra-arterial chemotherapy pump placement in patients with colorectal metastases to the liver. J Comput Assist Tomogr 2004;28 : 478-484[CrossRef][Medline]
  15. Kim HC, Kim TK, Sung KB, et al. Preoperative evaluation of hepatocellular carcinoma: combined use of CT with arterial portography and hepatic arteriography. AJR 2003;180 : 1593-1599[Abstract/Free Full Text]
  16. Tajima T, Honda H, Taguchi K, et al. Sequential hemodynamic change in hepatocellular carcinoma and dysplastic nodules: CT angiography and pathologic correlation. AJR 2002;178 : 885-897[Abstract/Free Full Text]
  17. Adachi B. Das Arteriensystem der Japaner, vol. Tokyo, Japan: Kenkyusha Press, 1928:11 -68
  18. Iida T, Hirata N, Hirakawa M, Noguchi T. Preoperative intraarterial infusion chemotherapy for advanced gastric cancer: a retrospective review of four cases. Radiat Med 2003;21 : 172-177[Medline]
  19. Suzuki S, Ueki T, Okamoto M, Fukuda M, Miyauchi T, Ishida M. Carcinoma of the gastric remnant fed by the accessory left gastric artery: a case report [in Japanese]. Nippon Geka Gakkai Zasshi1991; 92:89 -92[Medline]
  20. Tajima T, Honda H, Kuroiwa T, et al. Pulmonary complications after hepatic artery chemoembolization or infusion via the inferior phrenic artery for primary liver cancer. J Vasc Interv Radiol2002; 13:893 -900[Medline]
  21. Covey AM, Brody LA, Maluccio MA, Getrajdman GI, Brown KT. Variant hepatic arterial anatomy revisited: digital subtraction angiography performed in 600 patients. Radiology 2002;224 : 542-547[Abstract/Free Full Text]
  22. Miyaki T, Yamada M, Kumaki K. Aberrant course of the left gastric vein in the human: possibility of a persistent left portal vein. Acta Anat 1987;130 : 275-279[Medline]
  23. Ohashi I, Ina H, Hanafusa K, et al. Aberrant left gastric vein demonstrated by helical CT. J Comput Assist Tomogr1997; 21:996 -1000[CrossRef][Medline]
  24. Yoshimitsu K, Honda H, Kuroiwa T, et al. Unusual hemodynamics and pseudolesions of the noncirrhotic liver at CT. RadioGraphics 2001; 21 spec no:S81 -S96
  25. Kawata S, Murakami T, Kim T, et al. Multidetector CT: diagnostic impact of slice thickness on detection of hypervascular hepatocellular carcinoma. AJR 2002;179 : 61-66[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ishigami, K.
Right arrow Articles by Honda, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ishigami, K.
Right arrow Articles by Honda, H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS