YAO ET AL ORIGINAL ARTICLE Radiation Medicine: Vol. 21 No. 6, 252 257 p.p., 2003 Communicating Vein between the Left Renal Vein and Left Ascending Lumber Vein: Incidence and Significance on Abdominal CT Yuki Yao,* Yoshitaka Okada,** Minoru Yamato,* and Kuni Ohtomo*** Purpose: To examine the incidence and appearance of the communicating vein between the left renal vein (LRV) and the left ascending lumbar vein (LALV) on abdominal CT. Materials and Methods: We reviewed the appearance of the communicating vein on contrastenhanced CT obtained by multidetector-row CT (MDCT) scanner. One hundred patients without known abnormality in this region were randomly chosen and the following findings were recorded: (1) presence or absence of the visible communicating vein, (2) diameter of the communicating vein, (3) visible length of the vein (none, partial, or total), (4) laterality of the ascending lumbar vein, (5) distance between the superior mesenteric artery and the aorta. Results: The communicating vein was visible in 35 patients (35%). In 20 cases, this vein was visualized within the paraaortic region and could not be traced toward the LALV. The distance between the superior mesenteric artery and the aorta was narrower in the patients with visible communicating vein than in those without it, however, no statistically significant difference was demonstrated. Conclusion: This vein is commonly visible in the general population, and care should be taken not to confuse it with lymphadenopathy because this communicating vein was partially visualized within the paraaortic region in 20% of the cases. Key words: left renal vein, ascending lumber vein, communicating vein Received July 28, 2003; revision accepted October 3, 2003. *Department of Radiology, International Catholic Hospital **Department of Radiological Sciences, International University of Health and Welfare ***Department of Radiology, University of Tokyo Reprint requests to Yuki Yao, M.D., Department of Radiology, International Catholic Hospital, 2-5-1 Naka-ochiai, Shinjuku-ku, Tokyo 161-8521, JAPAN. INTRODUCTION HE VENOUS SYSTEM has a variety of communicating T branches, which may function as potential collateral pathways in case of venous obstruction. There are also various normal anatomic variants in these venous connections, which may create confusing radiologic appearances in some specific conditions. The communicating vein between the left renal vein (LRV) and the left ascending lumbar vein (LALV) is seen dorsal to the LRV and adjacent to the aorta (Fig. 1). Dilatation of this vein has been described as a potential diagnostic pitfall on abdominal computed tomography (CT), because this vein may mimic paraaortic lymphadenopathy on noncontrast CT or even on contrastenhanced CT when the veins are not sufficiently enhanced. 1,2 A full understanding of these retroperitoneal structures is imperative for retroperitoneal operations, and the surgeon should be informed of the presence of a dilated communicating vein, especially when the patient is a candidate for renal surgery, so that inadvertent bleeding can be avoided. On the other hand, exact evaluation of lymphadenopathy in the region of renal veins is required in the staging of patients with genitourinary neoplasm. Therefore, knowledge of the incidence and imaging features of the communicating vein between the LRV and LALV is of particular importance to avoid diffential diagnostic problems. Because of recent advances in multidetector-row CT (MDCT) scanners, the incidence of this communicating vein on CT images may be elevated, and, therefore, understanding of this diagnostic pitfall is more important. To our knowledge, however, there have been only a few case reports in the literature, and the prevalence and clinical significance of this finding have not been fully explored. 3 The purpose of this study was to elucidate the incidence and significance of this communicating 252 RADIATION MEDICINE
ORIGINAL ARTICLE vein in MDCT studies. MATERIALS AND METHODS Contrast-enhanced abdominal CT studies performed from May 2002 to December 2002 were retrospectively reviewed, and images of 100 patients were chosen for the study. These materials were randomly collected by using the terminal digits of the patients hospital ID numbers. Patients with renal or retroperitoneal abnormalities that could influence the venous anatomy were excluded from evaluation. CT images without sufficient opacification of the veins were also excluded. The patient population consisted of 42 men and 58 women with a mean age of 58 years (range, 22-79 years). In most patients, 100 ml of nonionic contrast medium at a concentration of 300 mgi/ml (Omnipaque 300; Daiichi Pharmaceutical Co., Ltd., Tokyo, Japan) was administered at a rate of 1.5-3.0 ml/sec thorough a 22- gauge IV cannula. As images in early phase were not included in this study analysis, the scan delay after contrast injection was typically set at 90 sec. All CT scans were obtained with an MDCT scanner (Asteion, Toshiba Medical Systems, Co., Ltd., Tochigi, Japan). Scanning was performed using a pitch of 5.5:1, 0.8-sec scanning time per rotation, and detector configuration of 4.0 3.0 mm. These parameters resulted in a table speed of 16.5 mm per rotation or 20.6 mm/sec. Other scanning parameters included 120-140 kvp and 200-240 ma. Images with a slice thickness of 7.0 mm were reconstructed with a 7.0 mm increment from the raw data. If a patient had undergone repeated CT examinations during his or her course, only one examination was used for evaluation. The appearance of the communicating vein between the LRV and LALV was evaluated by two radiologists (Y.Y., Y.O.), and a consensus was reached. Cases of enhanced vascular structure noted in the area between the LRV and LALV were chosen for further evaluation. An enhanced venous structure connected to both the LRV and venous channel to the LALV was recorded as a visible communicating vein. The course of the venous structure between the psoas major muscle and vertebra was considered to be a venous channel connected to the LALV. Presence or absence of the connection to the LRV and LALV was estimated respectively on several consecutive 7-mm-thick sections. To confirm these features, additional reconstruction of 3-mm thickness with 2-3-mm interval from the same raw data was performed in cases of equivocal findings, however, these 3-mm-thick images were not used for the following measurement. The largest diameter and length of the communicating vein were measured. The length of the Volume 21, Number 6 Fig. 1. Illustration of the cross-sectional anatomy of the communicating vein between the left renal vein (LRV) and left ascending lumbar vein (LALV). vein was categorized into three subgroups: (a) no visible vein; (b) a short segment of the vein within the paraaortic region (Fig. 2), which required 3-mm-thick reconstruction to visualize the channel to the LALV; and (c) a longer visible vein which can be traced until it reaches the space between the psoas major muscle and the vertebra (Fig. 3). Based on the assumption that a hemodynamic condition of the LRV or LALV may correlate with detection of the communicating vein, two indirect findings were also evaluated: (a) difference in size between left and right ascending lumbar veins (no difference, left-side dominant, or right-side dominant) and (b) the gap between the superior mesenteric artery (SMA) and the aorta. The gap was measured at the level where the LRV runs between these vessels. Comparison of categorical data was performed with the chi-square test or Fisher s exact test. Comparison of the gap between the SMA and the aorta was performed by unpaired t-test. Difference was considered significant if P was less than 0.05. RESULTS Direct findings (a) The communicating vein between the LRV and LALV was seen on contrast-enhanced CT studies in 35 patients (35%). There was no predominance in age, sex, or any specific underlying disease. The diameter of the largest part of these communicating veins ranged from 3 to 11 mm (mean, 6.7±1.9 mm). (b) In 20 patients (20%), only the paraaortic portion of the communicating vein was demonstrated. Reconstruction with 3-mm 253
YAO ET AL a b Fig. 2. Appearance of the visible short segment of the communicating vein between the LRV and LALV on continuous CT sections. The vein (arrow) is noted within the paraaortic region and cannot be traced into the space between the psoas muscle and vertebra. a b Fig. 3. Visible long segment of the communicating vein (arrow). The vein branches from the LRV and runs between the vertebral body and the psoas major muscle until it reaches the LALV. thickness was required in nine cases to confirm the connection to the LRV and in all 20 cases to confirm the connection to the LALV. (c) The long segment of the communicating vein to the connection with the LALV was visualized in 15 patients (15%). Six patients required 3-mm reconstruction to visualize the connection to the LRV. Indirect findings (a) The left and right ascending lumbar veins did not show any difference in size in 63 patients without a visible communicating vein (97%) and 30 patients with a visible communicating vein (86%). The left ascending lumbar vein was more prominent than the right in five patients with a visible communicating vein (14%). This finding was not seen in any patient without a visible communicating vein (P=0.023) (Table 1). (b) The gap between the proximal portion of the SMA and the aorta was narrower in the patients who had a visible communicating vein (11.6±5.2 mm) than in those without it (13.5±4.1 mm), however, the difference was not statistically significant (P=0.056) (Table 2, Fig. 4). DISCUSSION The ascending lumbar veins are paired and connect the common iliac vein, four pairs of lumbar veins, and iliolumbar veins. They ascend deep to the psoas muscle 254 RADIATION MEDICINE
ORIGINAL ARTICLE Table 1. Size difference between the right and left ascending lumbar veins in patients with and without visible communicating veins on CT Visible communicating vein Difference in the ascending lumbar veins Absent Present No difference 63 30 Right-side prominent 1 0 Left-side prominent 1 5 (P=0.0225) Note: Data are numbers of the patients in each subgroup. Table 2. Gap between the aorta and SMA in patients with and without visible communicating veins on CT Visible communicating vein Patients Absent Present P All patients (n=100) 13.5±4.1 mm 11.6±5.2 mm 0.056 (NS) Patients without marked dilatation of the left gonadal vein (n=94) 14.0±3.3 mm 11.8±4.1 mm 0.038 Note: Data are means ± standard deviation. NS = not significant. anterior to the medial portions of the lumbar transverse processes to join with the subcostal veins, which drain into the azygos vein on the right and the hemiazygos vein on the left, respectively. These junctures are usually under cover of the cranial extremity of the psoas major muscle and inferior to the head of the twelfth rib. 4 In autopsy study, complete exposure of the entire vessel was done in 81% of cadavers. The majority of interruptions occurred at the third lumbar level, however, the cranial portion was never absent. 4 The common location of the communicating vein between the LRV and LALV was around the second or the third lumbar level. During the developmental process of the inferior vena cava (IVC) and the azygos venous system, a pair of caudal cardinal veins (which almost disappear in developmental course), the subcardinal veins (which form the IVC and left gonadal vein), and the supracardinal veins (which mainly develop into the azygos venous system) form a vascular ring around the aorta with several transverse anastomotic channels 5 (Fig. 5). The LRV is usually formed from the transverse anastomotic vessel between the left and right subcardinal veins. Variations in this developmental process explain the formation of the retroaortic left renal vein, circumaortic renal venous ring, and the retrocaval ureter. 5-7 It is likely that the communicating vein between the LRV and LALV is a remnant of the anastomotic channels between the subcardinal vein and the supracardinal vein. The incidence of the communicating vein between the LRV and the ascending lumbar/ Volume 21, Number 6 Fig. 4. Scatter plot of the gap between the aorta and SMA in patients with and without visible communicating vein on CT images. hemiazygos venous system is approximately 90% in anatomic dissection. 4,8,9 And the LRV is bound down to the lateral vertebral border with this communicating vein and other collateral vessels in 60% of cases, which makes the LRV hard to liberate for surgeons. 4,8,9 In our study, contrast-enhanced CT demonstrated this 255
YAO ET AL connection in 35% of the patients. This disparity between the anatomic literature and radiologic findings may be attributed to the limited resolution of 7-mm-thick CT images, because the diameters of these communicating veins are not large enough to play a major hemodynamic role in most of the cases. Reconstruction with a thickness of 3 mm provided more detailed features of this vein. In 20 cases of short visible communicating vein, 3-mmthick reconstruction showed connection to the left renal vein that was not visualized on 7-mm-thick images. The incidence of visual communicating vein should be higher on 3-mm-thick images, however, we focused on the findings obtained with 7-mm-thick sections in this study. In the cases in which the communicating vein was detectable on CT scans, not only these congenital factors but hemodynamic conditions such as elevated left renal venous pressure might also play a role. It is well known that the LRV may be pinched between the aorta and SMA when the gap between these two vessels is narrow, resulting in elevated left renal venous pressure. In this series, the patients with a visible communicating vein on CT tended to have a narrower aorta-sma gap than those without it, however, the difference was not large enough to demonstrate statistical significance. Contrast-enhanced CT scan is a valuable procedure by which to search for retroperitoneal enlarged lymph nodes. It should be noted that demonstration of this communicating vein is not an uncommon finding on CT. Lien et al. 1 and Meanock et al. 2 reported that varicosity of this communicating vein mimics retroperitoneal lymphadenopathy and could be a potential diagnostic pitfall on abdominal CT. In our study, 20 of 35 cases with this communicating vein showed visible vessels confined in the paraaortic region, the feature of which mimics paraaortic lymphadenopathy, as neither the Fig. 5. Schematic drawing of the cross-sectional anatomy of the early developmental process of the venous system. The paired subcardinal veins form the IVC and the left gonadal vein. The LRV is formed from the transverse anastomotic vessel between the left and right subcardinal veins. The supracardinal veins develop into the azygos and hemiazygos vein. The caudal cardinal veins almost disappear in the later developmental course. connection to the LALV nor the full length of the communicating vein was confirmed on CT images. Enhancement on contrast-enhanced CT and the continuity on consecutive sections serve to distinguish these communicating veins from paraaortic lymphadenopathy (Fig. 6). However, because of the recent advances in MDCT, contrast-enhanced CT study of the abdomen is often completed before the communicating vein between Fig. 6. The communicating vein that mimics paraaortic lymphadenopathy. (a) Noncontrast and (b) early-phase contrast CT images show a nodular structure adjacent to the aorta, which mimics paraaortic lymphadenopathy (arrow). (c) This structure is clearly enhanced after intravenous contrast administration, showing the vascular nature of this varicose dilatation of the communicating vein. a b c 256 RADIATION MEDICINE
ORIGINAL ARTICLE the LRV and LALV is sufficiently enhanced. In such cases, knowledge of the anatomic and radiologic features of this communicating vein should be helpful to avoid problems in differentiation. Several limitations of this study need to be discussed. First, the technical details of the CT study were not completely uniform because of the retrospective nature of the study. The frequency of visualization of this communicating vein might be different according to the specific CT technique. We consider that the results of this study are applicable in most practical MDCT settings, however, the incidence and visualization of this communicating vein might be different when scanning is performed by a single-detector-row CT scanner. Second, reconstruction with other parameters from the same raw data was not performed in the cases without visual communicating vein. We focused on the CT findings on 7-mm-thick images because they are more relevant to the actual clinical setting, however, images reconstructed with thinner thickness may provide more frequent and clearer visualization of fine anatomic details such as this communicating vein and thus elevate the diagnostic accuracy. CONCLUSION In this study, the incidence of the communicating vein between the LRV and LALV was 35% on contrastenhanced abdominal CT images. Care should be taken not to confuse this vein with lymphadenopathy because this communicating vein was partially visualized within the paraaortic region in 20% of cases. Correlation between the aorta-sma gap and the incidence of visible communicating vein was not large enough to demonstrate statistical significance. REFERENCES 1) Lien HH, von Krogh J. Varicosity of the left renal ascending lumbar communicating vein: a pitfall in CT diagnosis. Radiology, 152: 484, 1984. 2) Meanock CI, Ward CS, Williams MP. The left ascending lumbar vein: a potential pitfall in CT diagnosis. Clin Radiol, 39: 565 566, 1988. 3) Picher JM, Padhani AR. Problems in diagnostic imaging: behind the left renal vein. Clin Anat, 10: 349 352, 1997. 4) Hoeltl W, Aharinejad S. Renal vein anatomy and its implication for retroperitoneal surgery. J Urol, 143: 1108 1114, 1990. 5) Sato T, Sato K. Regional anatomy for operative surgery of genitourinary organs. Inferior vena cava and left renal vein. Rinsho Hinyokika, 42: 1063 1074, 1988. (in Jpse.) 6) Trigaux JP, Vandroogenbroek S. Congenital anomalies of the inferior vena cava and left renal vein: evaluation with spiral CT. J Vasc Interv Radiol, 9: 339 345, 1998. 7) Royal SA, Callen PW. CT evaluation of anomalies of the inferior vena cava and left renal vein. AJR Am J Roentgenol, 132: 759 763, 1979. 8) Davis RA, Milloy FJ. Lumbar, renal and associated parietal and visceral veins based upon a study of 100 specimens. Surg Gynecol Obstet, 107: 1 22, 1958. 9) Pick JW, Anson BJ. The renal vascular pedicle. An anatomical study of 430 body-halves. J Urol, 44: 411 434, 1940. Volume 21, Number 6 257