Lymphatic leakage caused by lymphatic injury during abdominal surgery is a rare complication. In most cases, lymphatic leakage results from LN dissection and resolves spontaneously. However, hepatic lymphorrhea, which is leakage from the lymphatic ducts of the liver into the abdominal cavity, is an extremely rare complication, and only a few case reports have been published in the English literature [3,4,5]. In the present case, a massive amount of hepatic lymphorrhea occurred in the early postoperative period and was resistant to conservative therapies. Sclerotherapy following transhepatic lymphangiography proved to be effective in this case.
The lymphatic system in the peritoneal cavity is roughly divided into three parts: the lumbar, intestinal, and hepatic lymphatic systems [6]. There are two pathways in the hepatic lymphatic system: the ascending pathway along the hepatic veins, and the descending pathway along the hepatoduodenal ligament [7]. Nonchylous ascites consisting mainly of hepatic lymph is usually caused by injury to the lymphatic system from the hepatic hilum to the hepatoduodenal ligament. The liver normally produces a large amount of lymphatic fluid, and it has been estimated that typical lymphatic flow is approximately 0.25 mL/min [8]. Therefore, hepatic lymphorrhea can cause massive ascites. The most reliable surgical treatment for intractable hepatic lymphorrhea is ligation of the injured lymphatic vessels [9]. However, in clinical practice, it is difficult to identify the leakage site (injured lymphatic vessels) in the hepatoduodenal ligament, and only electrocautery of the suspected area of hepatic lymphorrhea or placement of a layer of fibrin glue can generally be performed [5].
It is well-known that the lymph flow is tenfold higher in cirrhotic patients than in healthy individuals due to comorbidities [9]. Even if the lymphatic duct in the hepatoduodenal ligament is damaged during surgery, the lymphatic flow is usually looped around the anastomotic branch, or an anastomosis of the lymphatic vascular channels occurs so that the damaged part is naturally occluded. Therefore, lymphatic leakage becomes problematic in cases of injury to the thoracic duct or cisterna chyli, as well as with the development of dilated lymphatic vessels around them. However, in the present case, pancreaticoduodenectomy did not extend into the thoracic duct and cisterna chyli. Moreover, the ascites was chylous, and triglyceride levels in the ascites were <110 mg/dL. Therefore, it was assumed that the cause of the massive lymphatic injury was not an injury to the thoracic duct and cisterna chyli, but to the lymphatic vessels in the hepatoduodenal ligament. Another factor suggesting hepatic lymphorrhea was that the lymphatic fluid oozed out markedly from the hepatoduodenal ligament during the surgery. In addition, the patient had chronic liver disease. Therefore, the hepatic lymphatic vessels were dilated, and the lymphatic flow was greater than that of normal liver. Thus, the surgically damaged lymphatic vessels were thought not to be spontaneously restored, and intractable hepatic lymphorrhea occurred. Considering these issues, dissection of the LN in the hepatoduodenal ligament in patients with cirrhosis should be performed with extreme caution.
In the present case, diagnostic percutaneous transhepatic lymphangiography was successfully performed, followed by insertion of a drainage catheter into the site of hepatic lymphorrhea. According to previous reports, hepatic lymphatic vessels can be demonstrated incidentally during percutaneous transhepatic cholangiography or portography [10, 11], since intrahepatic lymphatic vessels cannot be visualized by ultrasound, and they are difficult to puncture intentionally. Based on the anatomical information that lymphatic vessels run into Glisson’s sheath, the Chiba needle was advanced parallel to the portal vein and succeeded in puncturing the intrahepatic lymphatic vessels. Since these vessels show characteristic features, such as a beaded appearance and the presence of fine, lucent bands caused by valves, they are easy to recognize. In the present case, the hepatic lymphatic vessels may have been dilated due to cirrhosis. This may be one of the reasons why it was possible to visualize the intrahepatic lymphatic vessels in one session.
Matsumoto et al. [3] reported a case in which transhepatic lymphangiography was used to demonstrate the site of a hepatic lymphatic–peritoneal fistula in a patient who underwent gastrectomy with LN dissection for gastric cancer. They performed transhepatic lymphangiography in the same way as in the present case. They advanced the Chiba needle close to the right portal vein using ultrasound guidance. A 3-mL injection of indigo carmine liquid into the lymphatic vessels colored the ascites draining from the tube slightly blue, thus confirming the fistula. Guez et al. [4] reported a case of successful embolization of hepatic lymphorrhea using the Onyx Liquid Embolic System (Covidien, Plymouth, MN, USA).
In the present case, sclerotherapy was performed because the Chiba needle was spontaneously retracted, and it was difficult to embolize the intrahepatic lymphatic vessels. A single retrograde injection of OK-432 into the injured lymph vessels via a drainage catheter was unsuccessful due to the large amount of lymph flow. There have been several reports on the safety of the adhesion effect of OK-432 in closed spaces. Similar to the present case, the area imaged from the drainage tube was a closed space, and the distance from the anastomosis was sufficient, so OK-432 could be used with caution [12, 13]. Next, the obstructed area was gradually expanded by multiple injections of OK-432, and after seven injections, the leakage site was finally obstructed. There are a few case reports of hepatic lymphorrhea in the English literature [3,4,5], with approximately 20 cases of hepatic lymphorrhea as a complication of LN dissection in the hepatoduodenal ligament in patients undergoing gastric cancer resection in the Japanese literature [1].
To the best of our knowledge, this is the first report in which the site of hepatic lymphorrhea was visualized and successfully treated using retrograde injections of OK-432 into the injured lymphatic vessels.
It is important to be aware of the rare complications of hepatic lymphorrhea associated with injury to the hepatoduodenal ligament. In cases of persistent hepatic lymphorrhea, transhepatic lymphography followed by OK-432 injection might be a choice.