HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sheiman, R. G. Articles by Raptopoulos, V. Search for Related Content PubMed PubMed Citation Articles by Sheiman, R. G. Articles by Raptopoulos, V. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?

Alterations in Hepatic Perfusion Resulting from Splanchnic Venous Luminal Compromise Caused by Pancreatic Carcinoma

¹ All authors: Department of Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, 330 Brookline Ave., Boston, MA 02215.

Received June 17, 1999; accepted after revision November 15, 1999.

 
Address correspondence to R. G. Sheiman.

Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
OBJECTIVE. We determined whether alterations in hepatic enhancement exist on dual phase helical CT of the liver in patients with splanchnic venous luminal compromise resulting from pancreatic adenocarcinoma.

SUBJECTS AND METHODS. We examined the extent of hepatic enhancement on dual phase helical CT in 22 patients with pancreatic adenocarcinoma. Eleven patients had splanchnic venous luminal narrowing (flattening along at least 120° of the circumference) of the superior mesenteric vein with (n = 3) or without (n = 8) portal vein involvement caused by tumor. In the remaining patients, splanchnic vasculature appeared normal. An additional 16 patients without pancreatic or hepatic abnormality who underwent dual phase helical CT served as control subjects. We compared the extent of arterial phase and portal venous phase enhancement among the three groups.

RESULTS. The group of patients with splanchnic venous luminal compromise had significantly higher hepatic enhancement during the arterial phase (p < 0.01) and lower enhancement during the portal venous phase (p < 0.05) compared with the other two groups of patients. No significant difference in hepatic enhancement during either phase was noted between the control subjects and the patients with normal vasculature.

CONCLUSION. Because hepatic enhancement correlates with perfusion, splanchnic venous luminal compromise resulting from pancreatic adenocarcinoma likely causes decreased portal venous flow and compensatory increased hepatic arterial flow. This finding supports other evidence of a homeostatic mechanism that maintains hepatic perfusion.

Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
An increase in hepatic arterial flow with an associated decline in portal venous flow has been revealed on Doppler sonography as a response to metastatic disease in the liver [1,2,3]. These findings may be explained by the predominant arterial supply to hepatic metastases and by an increase in portal venous resistance, which may result from circulating humoral factors [4,5]. These findings have been supported by a more recent study in which patients with nonhepatic neoplasms, but with occult metastatic disease to the liver, showed increased hepatic parenchymal enhancement on helical CT during the arterial phase when compared with patients without hepatic involvement [6]. Somewhat similarly and also based on CT images of the liver, Miles et al. [7] revealed that hepatic arterial flow is increased in cirrhosis and portal venous flow is decreased, likely caused by portal hypertension. These studies suggest that hepatic enhancement patterns may be used as indicators of hepatic perfusion, and a physiologic, reciprocal relationship exists between hepatic arterial and portal venous flow.

In a study of patients with pancreatic adenocarcinoma involving the splanchnic vasculature [8], a significant decrease in portal vein enhancement was identified on helical CT images obtained 60 sec after the injection of IV contrast material when compared with control subjects and patients without involvement. Because the duration of contrast enhancement was greater than 60 sec (injection rate, 2 ml/sec; total contrast material dose, 2 ml/kg [120 ml]) and no differences in overall hepatic enhancement were observed, these authors postulated that hepatic arterial flow may increase as a response to decreased portal venous flow. If this hypothesis is true, it may represent another pathologic scenario in which attempted maintenance of hepatic blood supply exists through a reciprocal relationship between hepatic arterial and portal venous flow. Therefore, our goal was to use dual phase helical CT to prospectively assess any increase in the pure hepatic arterial enhancement or decrease in portal venous enhancement in patients with splanchnic venous luminal narrowing caused by pancreatic adenocarcinoma.

Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
From December 1997 to January 1999, 74 patients referred for dynamic helical CT for possible pancreatic adenocarcinoma were prospectively examined. All patients underwent dual phase helical CT (HiLight; General Electric Medical Systems, Milwaukee, WI) consisting of a hepatic arterial phase followed by a portal venous phase beginning at 22 sec and 70 sec, respectively, after the injection of contrast material. All patients fasted for at least 6 hr before undergoing CT. Imaging during the hepatic arterial phase started at the porta hepatis and continued through the pancreas. Portal venous phase imaging was performed from the top of the liver through the pancreas. CT parameters included a 3- to 5-mm slice thickness with a pitch of 1.3-1.7, amperage of 120 mA, and voltage of 200-240 kV. Initial unenhanced localizing images were obtained from the porta hepatis through the pancreas using a 7-mm slice thickness and a pitch of 2.0 All patients received IV ioversol (320 mg I/ml) (Optiray; Mallinckrodt, St. Louis, MO) administered with a power injector at a rate of 3.5 ml/sec. For all patients, the contrast material dose was tailored to the patient's weight at 2 ml/kg; however, the dose was never less than 100 ml and never more than 150 ml. This upper limit of contrast material dose was chosen because we wanted to obtain images in the hepatic arterial phase or portal venous phase. At an injection rate of 3.5 ml/sec, the maximum duration of injection was 43 sec, which insured that the hepatic artery was receiving little if any contrast material at the onset of imaging during the portal venous phase (70 sec).

Exclusion Criteria
Patients were eliminated from the study if they had superior mesenteric vein, portal vein, or splenic vein filling defects (thrombus on any axial CT images) or occlusion; adenopathy in the region of the porta hepatis; or any liver lesion, because the presence or absence of any of these conditions can alter hepatic arterial or splanchnic venous flow. For similar reasons, patients with mass effect on the splenic vein caused by a tumor or a history of cirrhosis, hepatitis, or cardiac or renal failure were also excluded. Arterial phase axial images for each patient were reviewed to ensure that the superior mesenteric artery and celiac axis were widely patent, confirming uncompromised hepatic arterial flow. When poststenotic dilatation or a stenosis of approximately 70% or greater in either vessel was suspected, images were retrospectively reconstructed with 50% overlap (1.5- to 2.5-mm slice thickness) and reviewed. If suspicion for a high-grade stenosis of either the superior mesenteric artery or celiac axis persisted, the patient was excluded. Overall, 36 patients were excluded, leaving 38 patients who met our selection criteria and formed the basis for the study.

Region-of-Interest Analysis
Each examination was sent to an Advantage Windows workstation (General Electric Medical Systems) at which hepatic attenuation measurements were obtained on three different unenhanced axial images and at corresponding levels on hepatic arterial phase and portal venous phase CT images. At least two attenuation measurements of hepatic parenchyma only were obtained from each lobe of the liver. Care was taken to exclude ducts and vessels. Region-of-interest size was variable but always greater than 100 mm² to minimize noise. Average hepatic attenuation values from unenhanced images and arterial phase and portal venous phase images were determined. Hepatic arterial phase enhancement and portal venous phase enhancement were then calculated by subtracting a patient's average hepatic attenuation on the unenhanced images from the hepatic attenuation on the arterial phase and portal venous phase images. All hepatic attenuation measurements were obtained prospectively without knowledge of the presence or absence of pancreatic abnormality.

Patients were then retrospectively divided into groups depending on the presence or absence of pancreatic adenocarcinoma. Patients with a normal liver and pancreas on helical CT and either a normal follow-up CT scan a minimum of 6 months later; normal ERCP within 2 weeks of their initial CT; or normal findings for liver function tests, amylase, and lipase values were considered free of any pancreatic and hepatic abnormality and considered control subjects (group 3: nine women and seven men; weight range, 50-108 kg; mean weight, 66 kg; age range, 36-85 years; mean age, 58.6 years). Patients with pancreatic adenocarcinoma (tumor confirmed by open surgical or imaging-guided biopsy or histologic evaluation of a surgically resected specimen) were divided into two groups according to whether an altered superior mesenteric vein contour and luminal compromise with or without involvement of the portal vein was present (group 1: six women and five men; weight range, 48-91 kg; mean weight, 64 kg; age range, 38-74 years; mean age, 65.5 years) or absent (group 2: five women and six men; weight range, 45-97 kg; mean weight, 63 kg; age range, 48-88 years; mean age, 71.5 years) on helical CT images obtained during the portal venous phase.

Splanchnic venous luminal compromise was defined as a loss of superior mesenteric vein concavity with flattening extending over at least 120° of vessel circumference (Fig. 1). This was not assumed to differentiate encased and nonencased vessels (no attempt was made to discriminate between these) but was chosen as a minimum threshold to discern patients who may or may not have altered splanchnic venous inflow to the liver. Patients in whom the criterion for splanchnic venous luminal compromise was met and involved the superior mesenteric veinportal vein confluence were also noted, but a more formal grading of the level of luminal compromise in the patients in group 1 beyond this was not performed. Hepatic arterial phase enhancement and portal venous phase enhancement values for each patient group were determined. Additionally, hepatic arterial phase enhancement divided by the sum of the hepatic arterial phase and portal venous phase enhancement, defined as the arterial enhancement contribution and thought to reflect the hepatic arterial component to total hepatic flow, was also calculated for each patient group. The values of these parameters were compared using the Newman-Keuls test (PROFIT System; Division of Research Resources, National Institutes of Health, Bethesda, MD) to discern any significant differences.

View larger version (131K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1. —68-year-old woman with pancreatic adenocarcinoma. CT scan reveals flattening along posterior aspect of superior mesenteric vein (arrow) caused by adenocarcinoma. Superimposed protractor shows luminal narrowing extending over 123° of vessel's circumference. Note adjacent biliary stent.

Hepatic enhancement during the arterial phase depends in part on the contrast material bolus reaching the hepatic arteries at the time of imaging (which, in turn, depends on patient weight and cardiac status). To ensure that any differences in hepatic arterial phase enhancement between our three patient groups were strictly the result of differences in hepatic arterial flow, the arterial phase enhancement value for each patient was normalized by dividing it by the patient's aortic attenuation measured at the level of the celiac axis on arterial phase images. Normalized values of hepatic arterial enhancement, which should adjust for differences in arterial opacification among patients, were also compared among patient groups.

Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Patient groups showed no significant difference with respect to age, sex, or weight. Mean hepatic attenuation on unenhanced images was 55.6, 56.5, and 54.5 H for groups 1, 2, and 3, respectively, and was not significantly different. Eight patients in group 1 revealed superior mesenteric vein luminal narrowing only, and three patients showed abnormal contour and narrowing at the superior mesenteric vein-portal vein confluence (Fig. 2). Table 1 summarizes the mean values for the arterial enhancement contribution and hepatic enhancement during the arterial phase and portal venous phase. Patients with splanchnic venous luminal compromise had significantly greater hepatic arterial phase enhancement compared with control subjects (p < 0.01) and patients with carcinoma but with normal splanchnic venous vasculature (p < 0.01). In the latter two groups, hepatic arterial phase enhancement was similar. This significant difference persisted even when normalizing hepatic arterial enhancement to adjust for differences in arterial enhancement among patients. Patients in group 1 also had significantly lower hepatic enhancement during the portal venous phase when compared with patients in groups 2 (p < 0.05) and 3 (p < 0.01). No significant difference in portal venous phase enhancement was identified between groups 2 and 3. In group 1, the hepatic arterial phase and portal venous phase enhancement values for three patients with luminal narrowing at the level of the superior mesenteric vein-portal vein confluence (mean, 14.0 ± 3.4 H and 56.2 ± 4.2 H, respectively) showed no difference compared with eight patients with superior mesenteric vein luminal narrowing only (mean, 20.2 ± 6.9 H and 59.66 ± 26.1 H, respectively).

View larger version (127K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2. —71-year-old man with pancreatic adenocarcinoma. CT scan (2.5-mm slice thickness) reveals narrowing of superior mesenteric vein lumen at level of confluence with portal vein. Note portal vein (arrow), which was better seen on more cephalad axial CT scans (not shown).

The arterial enhancement contribution, calculated by dividing hepatic arterial phase enhancement by the sum of hepatic arterial phase enhancement and portal phase enhancement, was significantly higher in patients with splanchnic venous narrowing than in patients with normal vessel appearance on CT (p < 0.01) and control subjects (p < 0.01). A scatterplot of patients' arterial enhancement contribution values by patient group showed no overlap of individual values in groups 1 and 2 (Fig. 3). Only one patient in group 1 had an enhancement contribution value that fell in the range of individual values for group 3. The mean arterial enhancement contribution value for patients with splanchnic luminal compromise was at least twice that of the two other patient groups.

View larger version (11K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3. —Graph shows arterial enhancement contribution values (AEC) for individual patients by patient group. In each group, horizontal lines delineate mean values and 95% confidence intervals. We noted no overlap of 95% confidence interval for individual AEC values for group 1 () compared with groups 2 () and 3 ().

Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Alterations in the hepatic blood supply resulting from intrinsic hepatic abnormalities have been detected on sonography and dynamic helical CT. In two studies, Leen et al. [1,2] used Doppler sonography to show that the arterial contribution to total hepatic perfusion is increased in patients with cirrhosis and metastatic disease of the liver, even when the metastatic disease was initially occult, and that total perfusion of the liver is decreased in patients with cirrhosis. Miles et al. [7] found similar results using nonhelical serial axial CT images of the liver but, more important, revealed that alterations in hepatic perfusion can be assessed using IV contrast material and evaluating changes in the hepatic enhancement over time. More recently, this principle was applied by Platt et al. [6], who assessed hepatic arterial enhancement among patients with known malignancy but without CT evidence of hepatic involvement. These investigators found arterial hepatic enhancement significantly higher in patients who developed CT-detectable hepatic metastases during an 18-month follow-up compared with patients who did not. The results of the study by Platt et al., which are substantiated by their earlier work [9], and those of Leen et al. [1,2] further confirm that alterations in hepatic perfusion can be detected through alterations in hepatic enhancement. Extending this concept a step further, Leggett et al. [10], using hepatic time-density curves, calculated and revealed that hepatic arterial perfusion (in milliliters per minute per milliliters of parenchyma) is increased and portal venous perfusion likely decreased in patients with overt colonic metastases to the liver.

Our findings reveal that patients with splanchnic luminal compromise resulting from pancreatic carcinoma have significantly higher hepatic enhancement during the arterial phase and significantly lower enhancement during the portal venous phase than control subjects and patients with carcinoma with normal splanchnic venous vasculature. By extrapolating from the works of Platt et al. [6,9] and Leggett et al. [10], hepatic arterial flow is increased and portal venous flow is decreased in patients with splanchnic venous luminal narrowing resulting from pancreatic adenocarcinoma. Our findings support a hepatic homeostatic mechanism to maintain total hepatic perfusion, which has been proposed by other researchers [5].

We made no attempt to determine a threshold value for any parameters studied that could be used to predict splanchnic venous involvement by pancreatic carcinoma and nonresectability. Splanchnic venous luminal narrowing was the minimum criterion used to place patients in group 1 because we believe that this would result in altered splanchnic venous flow to the liver. We made no attempt to distinguish patients who met this criterion because of direct tumor involvement from those who had a similar appearance of their splanchnic vasculature that resulted from, but was not involved by, adjacent tumor. Also, although to do so would have been potentially supportive of our data, we did not attempt to correlate hepatic arterial phase enhancement, portal venous phase enhancement, and the arterial enhancement contribution values with the extent of venous luminal compromise in our patients in group 1 because we are unaware of any formal guidelines for the quantitative assessment of venous luminal stenosis on CT images. Our primary goal was only to determine if splanchnic venous luminal narrowing decreased portal venous phase enhancement, and, if so, resulted in any compensatory increase in hepatic arterial phase enhancement, thereby supporting a homeostatic mechanism for maintaining total hepatic perfusion.

Leggett et al. [10] indicate that the use of hepatic enhancement measurements as an indicator of hepatic perfusion is crude because such measurements do not consider variations in the shape of the contrast material bolus within the aorta (the concentration of contrast material within the aortic lumen as a function of radial and z-axis position) caused by variations in patient cardiac output and total blood volume. We agreed and believed that our finding of significantly elevated arterial phase enhancement in patients with altered splanchnic venous contour by itself needed further substantiation. This is why hepatic arterial enhancement for each patient was normalized. Normalization, which considers variation in bolus shape, still revealed an increase in the hepatic arterial enhancement in patients in group 1. Additionally, we administered IV contrast material on the basis of patient weight rather than using a set dose, which also considers variations in a patient's total blood volume.

We believe that assessment of the celiac axis and superior mesenteric artery is required when attempting to correlate hepatic enhancement with hepatic perfusion on arterial phase CT images. No mention is made by Platt et al. [6,9] or Leggett et al. [10] of the status of the celiac axis or superior mesenteric artery in their patients. The patient populations of these studies are predominantly elderly, with mean ages of 54, 55, and 69.7 years, respectively; therefore, wide patency of these vessels cannot be assumed. By eliminating patients with suspected significant celiac or superior mesenteric artery stenosis and confirming vessel patency on arterial phase images, we avoided the possibility that differences in hepatic arterial enhancement may have resulted from intrinsic disease of these vessels.

Although we eliminated patients with evidence of hepatic metastases on CT and patients with a history of hepatic disease, we cannot be sure that some of our patients with pancreatic adenocarcinoma did not have hepatic micrometastases. As previously mentioned, hepatic arterial flow is increased when either overt liver metastases or micrometastases are present, which could impact our hepatic arterial phase enhancement values. However, only in cases of overt liver metastases have sonography and CT findings suggested a decline in portal venous flow, with this work substantiated by animal models [4]. Therefore, micrometastases could not explain the significant decrease in portal venous enhancement found in the patients of group 1. Also, the lack of significant difference of parameters examined among control subjects and patients with pancreatic carcinoma and normal splanchnic venous vasculature argues against micrometastases in this patient group.

Finally, we realize that other clinical scenarios related to pancreatic adenocarcinoma, such as porta hepatis adenopathy or thrombus within or occlusion of the superior mesenteric vein, portal vein, or splenic vein can result in a decline in venous inflow to the liver and possibly show findings similar to those of this study. However, we purposely chose to focus only on splanchnic venous luminal narrowing as a cause of compromised portal venous inflow to the liver because this scenario has been previously investigated [8] and is what prompted us to perform this study.

In conclusion, on dual phase helical CT, patients with splanchnic venous luminal compromise caused by pancreatic carcinoma have decreased hepatic enhancement during the portal venous phase and increased enhancement during the arterial phase. Because hepatic enhancement correlates with perfusion, our findings represent another pathologic scenario in which a reciprocal relationship exists between hepatic arterial and portal venous flow, supporting the existence of a homeostatic mechanism that maintains hepatic perfusion.

Acknowledgments
 
We thank Bernard Ransil for statistically evaluating our data.

References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Leen E, Goldberg JA, Anderson JR, et al. Hepatic perfusion changes in patients with liver metastases: comparison with those patients with cirrhosis. Gut 1993;34:554 -557[Abstract/Free Full Text]
2. Leen E, Angerson WJ, Wotherspoon H, Moule B, Cook TG, McArdle CS. Detection of colorectal liver metastases: comparison of laparotomy, CT, US, and Doppler perfusion index and evaluation of postoperative follow-up results. Radiology 1995;195:113 -116[Abstract/Free Full Text]
3. Leen E, Goldberg JA, Robertson J, et al. Detection of hepatic metastases using duplex/color Doppler sonography. Ann Surg 1991;214:599 -604[Medline]
4. Nott DM, Grime SJ, Yates J, et al. Changes in the hepatic perfusion index during the development of experimental hepatic tumours. Br J Surg 1989;76:259 -263[Medline]
5. Carter R, Anderson JH, Cooke TG, Baxter JN, Angerson WJ. Splanchnic blood flow changes in the presence of hepatic tumour: evidence of a humoral mediator. Br J Cancer 1994;69:1025 -1026[Medline]
6. Platt JF, Francis IR, Ellis JH, Reige KA. Liver metastases: early detection based on abnormal contrast material enhancement at dual-phase helical CT. Radiology 1997;205:49 -53[Abstract/Free Full Text]
7. Miles KA, Hayball MP, Dixon AK. Functional images of hepatic perfusion obtained with dynamic CT. Radiology 1993;188:405 -411[Abstract/Free Full Text]
8. Sheiman RG, Raptopoulos V. Delayed intravenous contrast medium washout from the small bowel in patients with pancreatic carcinoma and splanchnic venous invasion. J Comput Assist Tomogr 1996;20:924 -929[Medline]
9. Platt JF, Francis IR, Ellis JH, Reige KA. Difference in global hepatic enhancement assessed by dynamic CT in normal subjects and patients with hepatic metastases. J Comput Assist Tomogr 1997;21:348 -354[Medline]
10. Leggett DAC, Kelley BB, Bunce IH. Colorectal cancer: diagnostic potential of CT measurements of hepatic perfusion and implications for contrast enhancement protocols. Radiology 1997;205:716 -720[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sheiman, R. G. Articles by Raptopoulos, V. Search for Related Content PubMed PubMed Citation Articles by Sheiman, R. G. Articles by Raptopoulos, V. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?