In patients with hepatocellular carcinoma, massive peritoneal effusion is often used as an indication for contraindication or caution for surgery and intervention [1], while conservative treatment is very ineffective. However, some patients with hepatocellular carcinoma with high-flow APF and massive peritoneal effusion have been treated with aggressive interventions from July 2002 to August 2005. The results are analyzed and reported as follows.
1. Materials and methods
1.1 Clinical data
Nine patients, including seven males and two females, aged 39-62 years old, with an average of 54.1 years old. All cases were diagnosed with hepatocellular carcinoma according to the criteria announced by the National Conference on Liver Cancer in 1997, and all of them were combined with large amount of peritoneal effusion. Two cases of upper gastrointestinal hemorrhage had occurred recently. In all cases, DSA of the hepatic artery was performed and confirmed the presence of high-flow APF.
1.2 DSA manifestations
In all cases, high blood supply lesions were seen in the liver, but often the hepatic artery-portal vein shunt was not clearly visible due to the large number of shunts, while the portal vein trunk was seen to have retrohepatic flow, making the coronary vein of the stomach and even the splenic vein and mesenteric vein visible near the hepatic end. There were 7 cases of central shunts and 2 cases of peripheral shunts. 3 cases were combined with a small amount of hepatic artery-hepatic vein shunts, all of which were of peripheral type.
1.3 Interventional treatment method
For peripheral APF, a catheter was placed into the target artery after hepatic arteriography, as close to the fistula as possible. The gelatin sponge was made into granules about 2×2 mm in size, mixed with contrast agent, and injected slowly under fluoroscopy through the catheter to completely embolize the arterial branch. Then the tumor is embolized with iodinated oil.
For central APF, the catheter was placed as close to the fistula as possible, and for those who could not distinguish whether the fistula was located in the left or right hepatic artery, the catheter was stabilized in the intrinsic hepatic artery. The same method is used to prepare gelatin sponge pellets, and every time 3-5 pellets are injected, “smoke” is observed once for blood flow changes, and so on. When the arteriovenous shunt is significantly slowed down, the image will be repeated, and the location of the fistula and the blood supply to the tumor can often be further clarified. The catheter is then passed over or avoided the fistula to enter the target artery of the tumor, and the tumor is embolized with iodized oil as much as possible. The catheter is then placed proximal to the fistula and gelatin sponge pellets are continued until the hepatic artery-portal vein shunt is completely eliminated. Hepatic arteriogram was performed again, and if there was still high blood supply tumor foci, iodinated oil embolization was performed again. At the end of the procedure, indirect portal venography was performed.
During embolization, spring coils are not used as much as possible. In cases where the artery is not clearly visualized and the portal vein is almost simultaneously visualized with the artery, it is estimated that the fistula is large and the gelatin sponge particles may enter the portal vein through the fistula, so a 3 mm diameter steel ring with hair can be released near the fistula and then embolized with gelatin sponge.
After 3 to 4 weeks, the hepatic artery was re-imaged and embolization was performed again for residual APF and high blood supply tumor lesions as above.
2 .Results
After the first intervention: (1) APF completely disappeared in 7 cases, a small amount of shunt remained in 2 cases, and the direction of portal blood flow changed from retrohepatic to hepatic in all cases (Figure 3-6); (2) iodized oil was well deposited in the tumor lesions in 6 cases, and iodized oil was deposited in some of the lesions in 2 cases, and there were no obvious high blood supply lesions on the hepatic arteriogram at the end of the operation in all cases, and several small branches of the subpancreaticoduodenal artery supplied blood to the tumor in 1 case. In one case, there were several small branches of the inferior pancreaticoduodenal artery supplying blood to the tumor and a small amount of shunt to the portal vein, so embolization could not be performed; (3) within 2 weeks after surgery, ultrasonography showed that the peritoneal fluid disappeared completely in 5 cases and a small amount of residual peritoneal fluid in 4 cases; (4) after 2 weeks, liver function was reviewed and all 5 cases with significant impairment of liver function improved significantly; (5) in all cases, the original symptoms of abdominal distension, abdominal pain, anorexia and fatigue improved to different degrees.
After 3-4 weeks, the hepatic arteriogram again showed that: (1) APF completely disappeared in 3 cases, and the other 6 cases showed a small amount of shunt, but the portal vein was all hepatotropic; (2) the main trunk or large branches of the hepatic artery originally embolized by gelatin sponge had different degrees of recanalization, and the tumor lesion could be embolized through the recanalized artery; (3) the tumor lesion with high blood supply was visualized in 7 cases.
After re-intervention: (1) among the 6 cases showing a small amount of shunt, 4 cases showed complete disappearance of arteriovenous shunt; (2) among the 7 cases with high blood supply tumor lesions, 6 cases showed good deposition of iodinated oil in the lesions.
3. Discussion
APF has a high incidence in hepatocellular carcinoma, mostly seen in giant mass type and multi-blood supply hepatocellular carcinoma, which often indicates the severity of the lesion and is one of the important causes of portal hypertension in middle and late stage hepatocellular carcinoma; clinical manifestations of portal hypertension due to APF are dominated by peritoneal effusion and followed by bleeding from esophagogastric fundic varices [3]. High-flow APF is often an important cause of peritoneal effusion in hepatocellular carcinoma, which is often under-recognized.
Peritoneal effusion is a common and serious complication of hepatocellular carcinoma. Hepatocellular carcinoma combined with a large amount of peritoneal effusion is usually classified as advanced stage, thus losing the opportunity for surgery, and some scholars regard peritoneal effusion as a contraindication or caution indication for interventional treatment [1]. However, in practice, we found that some peritoneal effusions are caused by portal hypertension due to high-flow APF, and in such cases, if the hepatic artery-portal shunt is effectively blocked, the portal pressure can be reduced and thus the peritoneal effusion can be reduced or eliminated. CT-enhanced scans have high clinical value for the diagnosis of APF [4, 5] and can be used as a necessary tool for case selection. Therefore, in practice, it is necessary and necessary to distinguish between peritoneal effusion due to high-flow APF and other causes of peritoneal effusion, and the former should be treated with aggressive and effective interventions.
Arterial embolization is preferred for the treatment of APF [2, 6, 7]. embolization of APF, while minimizing the fractional flow, should take into account the compensatory capacity of liver function and hepatocellular carcinoma itself. Therefore, embolization of normal hepatic artery branches should be minimized during the interventional operation, and the catheter should be placed as close as possible to the fistula before embolization. However, for central type high-flow APF, there are multiple branches supplying blood to the fistula through the collateral circulation, and embolization in the main trunk of the hepatic innominate artery is often required.
Most authors choose spring coils as the primary embolization material [2, 6], which prevents damage to the portal vein behind the fistula. However, spring-ring embolization is a thicker vessel, which is permanently embolized, and it is easy to form collateral circulation after embolization, leading to recurrence of APF and complicating the blood supply to the tumor due to the obstruction of the main artery, which affects the subsequent arterial embolization chemotherapy. We chose 2×2mm homemade gelatin sponge particles as the main embolization material to avoid the above disadvantages. Of course, if the fistula is thick, the direct use of gelatin sponge pellets may embolize the portal vein, and then the blood flow can be slowed down by embolization with a spring ring first, and then supplemented by embolization with gelatin sponge. However, we did not encounter such a large fistula in practice, and all gelatin sponges were applied without portal vein embolization.
In patients with hepatocellular carcinoma with high-flow APF, APF may not be the only cause of peritoneal effusion, but blocking the hepatic artery-portal vein shunt and reducing portal vein pressure may still achieve good results. In one of the cases in this group, abdominal pain, fever, abdominal muscle tension, bloody and cloudy peritoneal fluid, a large number of nucleated cells and cancer cells were detected, and primary peritonitis and abdominal metastasis were considered.
Since this intervention has the advantages of small trauma, simplicity and safety, and it works quickly after eliminating the shunt to reduce the portal pressure, it can also be applied to patients with large amount of peritoneal fluid, cachexia and weak constitution. At this time, the intervention should be simple and fast, not seeking perfection, in order to effectively reduce the portal vein pressure, and it is possible to achieve the desired effect. In these cases, the large amount of postoperative peritoneal fluid is rapidly absorbed and the cardiac load is increased, which may aggravate the already existing cardiac insufficiency, so clinical work should pay attention to the corresponding treatment. High-flow APF is often combined with portal vein aneurysm embolism, and interventional treatment mostly requires complete embolization of the intrinsic hepatic artery or the main branches of the left and right hepatic arteries. Therefore, the patency of the portal venous system must be evaluated before embolization treatment, and the blood supply to the liver after arterial embolization must be assessed to prevent ischemic necrosis of the liver, reduced liver function and accelerated deterioration of the disease.
In conclusion, a significant proportion of cases with advanced hepatocellular carcinoma complicated by peritoneal effusion have high-flow APF, which can achieve significant efficacy after active and appropriate interventional treatment. For such cases, we should be good at detecting and actively perform interventional treatment in clinical work.
Figure 1 Common hepatic arteriogram showing central high-flow APF with early visualization of portal vein in the artery.
Figure 2 Main portal vein flowing against the liver with open side branches.
Figure 3 Embolization of the hepatic artery with 2×2 mm gelatin sponge particles to eliminate APF.
Figure 4 The ectopic hepatic artery originating from the superior mesenteric artery still has a small APF, and the intrahepatic cancer foci are clearly visualized.
Figure 5 The small APF was embolized with gelatin sponge particles by super-selective cannulation first, and then the intrahepatic cancer foci were embolized with iodinated oil, and the re-imaging showed no APF and high blood supply cancer foci.
Figure 6 Indirect portal vein angiography shows portal blood flow to the liver.