| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Abdominal Imaging, Department of Radiology, University of
Pittsburgh Medical Center, 200 Lothrop St., Pittsburgh, PA 15213.
2 Department of Radiology, Western Pennsylvania Hospital, 4800 Friendship Ave.,
Pittsburgh, PA 15224.
3 Department of Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA
15213.
Received October 21, 2002;
accepted after revision January 30, 2003.
Address correspondence to V. Kapoor.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed the medical records of 26 patients with hepatic colorectal metastases who had been scheduled for implantation of a hepatic artery pump. Before surgery, all patients underwent MDCT arteriography with volumetric 3D rendering of the hepatic vessels. The axial and 3D arteriograms were evaluated for their usefulness in depicting hepatic arterial anatomy. Subsequently, three patients also underwent catheter angiography. Twenty-two of the 26 patients imaged had a hepatic artery floxuridine infusion pump implanted. Results of the CT arteriography were correlated with findings at surgery or on catheter angiography if surgery was not performed.
RESULTS. MDCT arteriography correctly revealed hepatic arterial anatomy in all 25 patients with angiographic or surgical confirmation. One patient with aberrant hepatic arterial anatomy did not have angiographic or surgical confirmation. Classic hepatic arterial anatomy was identified in 16 (64%) of 25 patients. The following hepatic arterial variants were found in one patient each: the common hepatic artery arising directly from the aorta; a replaced left hepatic artery; an accessory right hepatic artery; a replaced left hepatic artery and accessory right hepatic artery; a replaced right hepatic artery; a right hepatic arterial branch arising early (before the origin of the gastroduodenal artery); and replaced right and left hepatic arteries. Three patients were not suitable candidates for placement of a hepatic artery floxuridine pump. The patient who had no angiographic or surgical confirmation was also not considered a good surgical candidate because of replaced right and left hepatic arteries. Two patients (8%) had an accessory left hepatic artery.
CONCLUSION. MDCT arteriography with volumetric 3D rendering is an accurate, noninvasive method of depicting hepatic arterial anatomy and, therefore, of selecting patients with colorectal metastatic disease who could benefit from hepatic artery pump implantation. Catheter angiography provides no additional information, and we have eliminated it as a routine preoperative imaging examination.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 5-year survival rate among patients who receive no treatment is approximately 3% [3]. Targeted delivery of chemotherapy to liver metastases via a surgically placed hepatic artery infusion pump has been shown to produce improved tumor response, which suggests that such treatment extends survival [4]. Multidetector (MDCT) arteriography with volumetric three-dimensional (3D) rendering has been studied extensively as a method for depicting hepatic and splenic arterial anatomy [5–7]. Three-dimensional depiction of vascular and visceral interrelationships has the potential to assist surgeons in planning therapy. To our knowledge, application of MDCT arteriography in the evaluation of patients for hepatic arterial chemotherapy has not been previously reported.
The purpose of our study was to determine whether CT arteriography with volumetric 3D rendering could replace digital subtraction angiography for the preoperative evaluation of candidate suitability and planning for the placement of a hepatic artery floxuridine infusion pump in patients with metastatic colorectal disease.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We classified the anatomy of the hepatic arterial supply as either classic or variant (Michels' classification) [8]. Twenty-two of the 26 patients studied had a hepatic artery floxuridine pump implanted. The volumetric 3D-rendered images were available to the surgeons in the operating room at the time of the surgery. The results of MDCT arteriography were correlated with surgical findings or with catheter angiograms if surgery was not performed.
MDCT Technique
All MDCT arteriograms were performed on a LightSpeed QX/i CT scanner
(General Electric Medical Systems, Milwaukee, WI). Scans were obtained in
three phases: unenhanced, hepatic arterial, and portal venous phases. After
the unenhanced scan of the liver was acquired, we placed an 18- to 20-gauge
plastic IV catheter for hepatic arteriography in an antecubital vein and
connected the catheter to a power injector. The scanning time delay was
determined by administering a 20-mL timing bolus of iodinated contrast
material (Optiray 350, Mallinckrodt, St. Louis, MO) delivered at a rate of
4–5 mL/sec. A single-level (zero-space interval) 1-sec scanning time and
a 5-mm slice thickness were used to obtain scans at the level of the origin of
the celiac artery. After a 10-sec delay, scanning was initiated, and scans
were acquired every 2 sec for 30 sec. A region-of-interest cursor was placed
over the abdominal aorta, and the change in abdominal aortic enhancement over
time was measured, thereby providing a graph of enhancement (in Hounsfield
units) versus time (seconds). The peak of the graph (time required to reach
peak enhancement) indicated the delay for the arterial phase scanning.
All patients were asked to drink 8 oz (0.237 mL) of water just before scanning began so that the difference in attenuation between the gastroduodenal region and the enhanced vessels would be accentuated. All patients then received 150 mL of low-osmolarity iodinated contrast material (Optiray 350) injected IV at a rate of 4–5 mL/sec using a power injector (Medrad, Indianola, PA) for the arterial phase of the examination. For this phase of imaging, we used a four-detector configuration and an interslice gap of 1.25 mm in high-speed mode. The table speed was 7.5 mm per one tube rotation of 0.8 sec. The pitch in the high-speed mode was 6.0 (7.5 mm divided by 1.25 mm).
Scanning began at the level of the dome of the liver and covered the entire length of the liver (as determined by the unenhanced scans). Depending on the size of the liver, the arterial phase could be completed in 15–22 sec during a single breath-hold. After an interscan delay of 5 sec for table repositioning, the portal venous phase of the examination was performed. The MDCT arterial phase data were retrospectively reconstructed at 1.25-mm thickness with no overlap, and volumetric 3D rendering was then performed on a workstation (Advantage Windows 3.1 or 4.0, General Electric Medical Systems). Two experienced radiologists combined the axial and 3D images to reach a consensus on each patient's hepatic vascular anatomy before surgery or catheter angiography. Performance of the volumetric 3D rendering at the workstation required 10–20 min. The time required varied according to the complexity of the hepatic arterial anatomy as well as the level of experience of the radiologist or CT technologist performing the procedure.
Guidelines for Pump Insertion
Most of the patients referred for insertion of a hepatic artery infusion
pump had unresectable colorectal disease that had metastasized to the liver,
and most had previously undergone systemic chemotherapy that had failed to
elicit tumor response. All patients received either a 20- or 35-mL implantable
continuous chemotherapy pump (Isomed, Medtronic, Minneapolis, MN);
cholecystectomy was performed if it had not been done previously. The pump
catheter tubing was placed into the gastroduodenal artery with the tip of the
tubing at the junction of the gastroduodenal and hepatic arteries. Before the
pump was inserted, a 5-cm zone in the gastroduodenal region was devascularized
via ligation of the right gastric artery, any accessory hepatic artery
branches, and the gastroduodenal artery distal relative to the site of the
pump. At the conclusion of the procedure, fluorescein sodium dye was injected
through the infusion catheter to rule out the presence of leaks and to confirm
that the arterial supply to both lobes of the liver was adequate.
The volumetric 3D-rendered images were available to the surgeon in the operating room at the time of the surgery. We used the findings at surgery or on catheter angiography (in patients who did not undergo surgery) as the gold standard with which to correlate the results of the volumetric 3D-rendered imaging.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Michels' classification describes nine types of variants in hepatic arterial anatomy. We found examples of several of these variants among our patients: type II, a replaced left hepatic artery (one patient); type III, a replaced right hepatic artery (one patient); type IV, replaced right and left hepatic arteries (one patient); type V, an accessory left hepatic artery (two patients); type VI, an accessory right hepatic artery (one patient); type VII, accessory right and left hepatic arteries (0 patients); type VIII, a replaced right or left hepatic artery with the other hepatic artery being an accessory artery (one patient with a replaced left hepatic artery and an accessory right hepatic artery); and type IX, the hepatic trunk acting as a branch of the superior mesenteric artery (0 patients). In addition, we found one patient in whom the common hepatic artery arose directly from the aorta and one patient in whom the right hepatic arterial branch arose early (Fig. 3A, 3B), before the origin of the gastroduodenal artery.
|
|
Three patients were judged by the surgical team to be unsuitable candidates for placement of a hepatic artery floxuridine pump: the patients with Michels' type III (Fig. 4A, 4B, 4C, 4D) and Michels' type IV variants as well as the patient with the early-arising right hepatic arterial branch. The patient with no angiographic or surgical confirmation had replaced right and left hepatic arteries and was also not considered a surgical candidate. Five of the 22 patients who had the chemoinfusion pump insertion also underwent radiofrequency ablation of hepatic metastases, and four patients had resection of a hepatic mass at the time of pump placement. Because the MDCT angiography provided all the anatomic information needed to plan for placement of the hepatic arterial pump catheter, catheter angiography was eliminated as a routine preoperative imaging study.
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hepatic arterial chemotherapy infusion may be effective in patients with both resectable and unresectable liver metastasis. The liver has a dual blood supply, with normal hepatocytes deriving their blood supply primarily from the portal vein (and to a lesser extent from the hepatic artery), whereas metastatic lesions derive their blood supply primarily from the hepatic artery [11]. Hepatic arterial infusion delivers maximal doses of the chemotherapeutic agents to the hepatic malignancy while exposing the normal liver tissue and other organs to a relatively reduced amount of the agents. After the resectioning of clinically evident metastases, "micrometastases" measuring 1–3 mm may remain undetected. Subsequent hepatic arterial chemotherapy is provided to treat and potentially to eliminate these small lesions.
Direct infusion of a chemotherapeutic drug into the hepatic artery exposes the metastases to high concentrations of chemotherapeutic drugs while relatively sparing the normal liver tissue [12]. Response rates in patients with resectable liver metastases who are treated with hepatic artery infusion and systemic chemotherapy are better than the response rates in patients who are treated with systemic chemotherapy alone [10].
Digital subtraction angiography has been used for the depiction of hepatic vascular anatomy in patients undergoing liver surgery. The advent of MDCT has made it possible to perform CT arteriography of the abdominal vessels. MDCT arteriography with volumetric 3D-rendered imaging has been studied for its usefulness in depicting hepatic and splenic arterial anatomy [5–7]. This technique is used extensively for the preoperative selection of living related liver donors as well as for the evaluation of the vascular anatomy of the recipients [13, 14]. We studied the usefulness of a similar technique that displays abdominal arterial anatomy and provides valuable information about the extent of intraabdominal disease in patients being evaluated as possible candidates for placement of hepatic arterial floxuridine infusion pump.
Michels [8] described hepatic arterial anatomy and its variations in detail in 1955, using the results of cadaveric dissection. We have used the same classification. Conventional hepatic arterial anatomy (Michels' type I) is described as arterial anatomy in which the common hepatic artery arises from the celiac artery and then divides into the proper hepatic artery and the gastroduodenal artery. The right and left hepatic arteries arise from the proper hepatic artery. The middle hepatic artery, which supplies Couinaud segment IV, may arise from the right, left, or proper hepatic artery. This pattern was the most common arterial anatomy encountered in our study, found in 64% of our patients. Patients with this arterial pattern were suitable candidates for insertion of a hepatic arterial floxuridine infusion pump.
Burke et al. [15] showed that the response to intraarterial infusion in patients with normal hepatic arterial anatomy was better than that in patients with aberrant arteries. Burke postulated that the difference in response was caused by the ineffectiveness of intrahepatic collateral circulation in delivering the chemotherapeutic drug compared with effectiveness of delivery possible through the main hepatic artery.
The patterns of hepatic arterial anatomy that are considered marginal for insertion of a hepatic arterial infusion pump are a replaced left hepatic artery, a replaced right hepatic artery, or replaced right and left hepatic arteries (Michels' class type II–IV). Patients with a replaced left hepatic artery (Michels' type II) may still qualify for a pump placement if the metastases are confined to or are predominantly found in the right lobe of the liver. A patient with a colorectal metastases located predominantly in the left hepatic lobe may be disqualified as a candidate for pump placement if the gastroduodenal artery arises from the right hepatic artery after the bifurcation of the proper hepatic artery or from a replaced right hepatic artery. This anatomic arrangement has been described as a type B variant by Daly et al. [16], who modified the Michels classifications.
Of the four patients in our study not deemed to be suitable candidates for hepatic arterial pump insertion, two had replaced right and left hepatic arteries (Michels' type IV), with the proper hepatic artery dividing into the gastroduodenal artery and the middle hepatic artery (Fig. 4A, 4B, 4C, 4D). A pump inserted into the gastroduodenal artery in these patients would have supplied only segment IV. The third unsuitable candidate for pump placement had a replaced right hepatic artery (Michels' type III), with the gastroduodenal artery arising from the left hepatic artery (Fig. 5A, 5B). The remaining patient had an early right hepatic arterial branch that arose before the origin of the gastroduodenal artery, which originated from the left hepatic artery (Fig. 3A, 3B). A pump placed in either of these two patients would have supplied only the left liver lobe. An alternative in patients with accessory or replaced hepatic arteries and bilobar metastases is either insertion of dual-infusion catheters into main and aberrant arteries [15] or ligation of the smaller nondominant hepatic artery.
|
|
Rates for pump-related complications reported in the literature range from 28.8% to 60.0% [17–18]. The complications may be related to misperfusion (extrahepatic or intrahepatic), to direct chemotoxicity of the agent being infused, or to surgery (including failure of infusion equipment). Complications occurred in six (27%) of the 22 patients in our series who had a floxuridine infusion pump insertion. Misperfusion occurred in three patients (13.6%), surgery-related complications occurred in two patients (9%), and chemical cholangitis occurred 11 months after pump placement in one patient (4.5%).
Early postoperative complications (those appearing within 30 days of the surgery) occurred in three patients in our study. In two of these patients, the complications necessitated the removal of the pump: one patient had dissection of the hepatic artery, and the other developed deep venous thrombosis of the lower extremity that required anticoagulation therapy that resulted in intraperitoneal bleeding (Fig. 6) adjacent to the insertion of the infusion catheter into the gastroduodenal artery. Although no leak at the site could be found at exploratory laparotomy, the bleeding stopped after removal of the pump. Another patient developed duodenitis approximately 14 weeks after insertion of the pump. The duodenum was being supplied by a small duodenal artery that originated between the tip of the infusion catheter and the junction of the gastroduodenal artery with the proper hepatic artery (Fig. 7A, 7B, 7C, 7D, 7E, 7F). This small vessel could not be visualized on MDCT angiography, even on retrospective analysis of the images. The patient was successfully treated with endovascular coil occlusion of the vessel. Two other patients also had complications related to extrahepatic misperfusion that were treated successfully with embolization (Fig. 8A, 8B, 8C, 8D).
|
|
|
|
|
|
|
|
|
|
|
Other complications associated with the hepatic arterial chemoinfusion pump that have been reported in the literature are hematoma, seroma, or abscess formation at the site of the pump pocket; hepatic artery thrombosis or aneurysm; gastric and duodenal inflammation or ulceration; chemical hepatitis; sclerosing cholangitis; and pump malfunction [16–19]. Complications related to misperfusion can be minimized by evaluating hepatic and extrahepatic organ perfusion with postoperative infusion pump scintigraphy using 99mtechnetium-macroaggregated albumin [20]. The surgeons at our institution order 99mtechnetium-macroaggregated albumin pump studies only in patients believed to have misperfusion. Accessing the pump pocket with a needle during the early postoperative period, when a pump-pocket seroma is invariably present, may increase the risk of iatrogenic infection or bleeding.
At our institution, hepatic MDCT arteriography with volumetric 3D rendering has almost completely replaced digital subtraction angiography as a method of preoperatively evaluating candidates for placement of a hepatic arterial floxuridine infusion pump. MDCT arteriography has many advantages over catheter angiography. CT is used for the initial evaluation and follow-up of most patients with hepatic metastases, and modifying the CT scanning technique permits adequate collection of arterial information, thereby avoiding an invasive and expensive examination. CT also provides valuable information about the number, size, and distribution of hepatic metastases and about the presence and extent of extrahepatic disease. We now reserve use of catheter angiography for patients in whom angiographic therapeutic intervention may be required. The only remaining indication for catheter angiography is a technically inadequate MDCT angiogram, usually caused by a lack of adequate IV access, the extreme obesity of a patient, or an equipment malfunction, but these are uncommon occurrences. Although the small number of patients limits our study, the incidence of complications related to misperfusion in our series is comparable to that described in the literature [17].
In conclusion, we found that surgical placement of an infusion pump for hepatic arterial infusion of chemotherapy provides survival benefits in selected patients with metastatic colorectal cancer. High-quality angiograms can be constructed by computer modifications of the MDCT scans, yielding accurate 3D images of hepatic arterial anatomy that obviate invasive catheter angiography in potential candidates for hepatic arterial chemotherapy. However, catheter angiography continues to play an important role in the management of postoperative complications related to hepatic arterial chemotherapy misperfusion.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Shimada, H. Isoda, T. Okada, T. Kamae, S. Arizono, Y. Hirokawa, and K. Togashi Unenhanced MR Portography With a Half-Fourier Fast Spin-Echo Sequence and Time-Space Labeling Inversion Pulses: Preliminary Results Am. J. Roentgenol., July 1, 2009; 193(1): 106 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. B. Winston, N. A. Lee, W. R. Jarnagin, J. Teitcher, R. P. DeMatteo, Y. Fong, and L. H. Blumgart CT Angiography for Delineation of Celiac and Superior Mesenteric Artery Variants in Patients Undergoing Hepatobiliary and Pancreatic Surgery Am. J. Roentgenol., July 1, 2007; 189(1): W13 - W19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ishigami, K. Yoshimitsu, H. Irie, T. Tajima, Y. Asayama, M. Hirakawa, and H. Honda Accessory Left Gastric Artery from Left Hepatic Artery Shown on MDCT and Conventional Angiography: Correlation with CT Hepatic Arteriography Am. J. Roentgenol., October 1, 2006; 187(4): 1002 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. T. Johnson and E. K. Fishman IV Contrast Selection for MDCT: Current Thoughts and Practice Am. J. Roentgenol., February 1, 2006; 186(2): 406 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. B. Hunter and M. S. Taljanovic Medical Devices of the Abdomen and Pelvis RadioGraphics, March 1, 2005; 25(2): 503 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Stemmler, E. K. Paulson, F. J. Thornton, S. R. Winters, R. C. Nelson, and B. M. Clary Dual-Phase 3D MDCT Angiography for Evaluation of the Liver Before Hepatic Resection Am. J. Roentgenol., December 1, 2004; 183(6): 1551 - 1557. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
